id stringlengths 24 24 | question_id stringlengths 24 24 | axis stringclasses 4
values | num_turns int64 3 19 | messages listlengths 3 19 | target_question stringlengths 18 692 | pass_criteria stringclasses 1
value | target_question_en stringlengths 50 1.28k | messages_en listlengths 3 19 | translation_metadata dict |
|---|---|---|---|---|---|---|---|---|---|
674552683acc22154b07a598 | 674552683acc22154b07a598 | INFERENCE_MEMORY | 3 | [
{
"content": "안녕하세요! 저는 유엔 본부에서 근무하는 국제 관계 전문가예요. 제 업무 특성상 여러 나라의 외교관들을 지속적으로 만나야 해요. 저는 뉴욕에서 택시나 대중교통을 이용하는 것을 싫어해요. 유엔 본부에서 도보 5분 이내 거리에 있는 장소를 선호해요.",
"role": "user"
},
{
"content": "안녕하세요! 뉴욕 유엔 본부 근처의 장소를 찾고 계시다니 반갑습니다. 도보 5분 이내에 있는 몇 가지 장소를 추천해 드리겠습니다:\n\n1. 다그 함마르셸드 플라자(Dag Hammarskjold Plaza... | 선택된 레스토랑들이 유엔 본부에서 도보 5분 이내 거리에 있습니까? | YES | Are the restaurants chosen within a 5-minute walk from the UN headquarters? | [
{
"content": "Hello! I am an International relations expert working at the UN headquarters. My work requires me to consistently meet with diplomats from various countries. I hate using taxis or public transportation in New York. I prefer venues that are within a 5-minute walk from the UN headquarters.",
"r... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674552684d7f0f0dad442da6 | 674552684d7f0f0dad442da6 | SELF_COHERENCE | 9 | [
{
"content": "일본 두루미와 그들의 매혹적인 구애 춤에 대해 공부하고 있어요. 일본에서 문화적으로 매우 중요한 의미를 지닌다고 알고 있는데, 그들의 상징성과 이를 기념하는 축제나 행사에 대해 더 알려주실 수 있나요?",
"role": "user"
},
{
"content": "일본 두루미(일본어로 탄초)는 실제로 일본 문화와 상징성에 깊이 스며들어 있습니다. 이들은 평생 한 마리의 배우자와 짝을 이루고 수십 년을 살 수 있기 때문에 장수, 정절, 그리고 행운의 상징으로 여겨집니다.\n\n일본 문화에서 이 두루미는 다양한 ... | 답변에서 쿠시로 탄초 축제의 가장 좋은 시기가 보통 2월 말이라고 명시하고 있습니까? | YES | Does the response state that the best time for the Tancho Festival in Kushiro is typically in late February? | [
{
"content": "I've been studying the Japanese red-crowned cranes and their fascinating mating dances. I understand they hold significant cultural importance in Japan. Can you tell me more about their symbolism and any festivals or events that celebrate them?",
"role": "user"
},
{
"content": "The Jap... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745526875828b24787b636f | 6745526875828b24787b636f | INSTRUCTION_RETENTION | 7 | [
{
"content": "안녕하세요, 지역 모임에서 가족의 역사적 중요성에 대해 말씀해 주실 할아버님을 모시고 싶어 하는 이벤트 기획자입니다. 대화 전체 동안 '당신'이나 '우리'와 같은 인칭 대명사를 사용하지 말고, 어디에서도 굵은 글씨를 사용하지 말아 주세요.",
"role": "user"
},
{
"content": "연락해 주셔서 감사합니다. 오해가 있는 것 같습니다. AI 어시스턴트로서, 이벤트에서 공유할 할아버지나 가족사가 존재하지 않습니다. AI는 지역 모임을 위한 다른 잠재적 연사나 주제를 탐색하는 것을 돕거나, 다양... | 모델이 계보학자의 자격 및 인증, 소프트웨어 및 도구, 계보학자에 대한 수요, 그리고 계보학의 미래를 논의하는 시각적으로 보기 좋은 응답을 제시할 때 굵은 글씨를 사용하지 않아야 한다는 점을 기억했습니까? | YES | Did the model recall to avoid using bold font when presenting a visually appealing response discussing the qualifications and certifications, the software and tools, the demand for genealogists, and the future of genealogy? | [
{
"content": "Hello, I'm an event organizer interested in having your grandfather speak at our local gatherings about his family's historical significance. For the entire conversation, please do not use personal pronouns such as 'you' or 'we,' and do not use boldface anywhere.",
"role": "user"
},
{
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455268a969fda3803801e9 | 67455268a969fda3803801e9 | INSTRUCTION_RETENTION | 5 | [
{
"content": "다가오는 레슬링 이벤트를 기획 중인데, 전문가님의 오랜 경험을 바탕으로 이벤트 기획을 도와주셨으면 해요. 첫 번째 질문은 무엇이 경기를 기억에 남게 만드나요? 제가 묻는 각 질문에 대해 딱 2문장으로만 답변해 주세요. 각 문장은 15단어 미만이어야 해요.",
"role": "user"
},
{
"content": "경기는 높은 이해관계가 걸린 스토리텔링과 인상적인 운동 능력의 결합을 통해 기억에 남게 됩니다. 예측 불가능한 결과와 레슬러들 사이의 정서적 유대감 또한 경기의 기억에 남는 요소가 됩니다.\n\n... | 모델이 두 문장으로 답변하고, 두 문장 모두 15단어 이하를 사용해야 한다는 규칙을 기억하고 있나요? | YES | Does the model remember to respond with two sentences, both of which use 15 words or less? | [
{
"content": "I am organizing an upcoming wrestling event, and would like some help from someone with your years of expertise to help me with my event organizing. The first question I have is what makes a match memorable? Please response to each question I ask using only 2 sentences. Keep each sentence under 15... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455268bc2ba9e69b8618f8 | 67455268bc2ba9e69b8618f8 | INSTRUCTION_RETENTION | 3 | [
{
"content": "사회학적 이슈에 대해 몇 가지 질문이 있어요. 이러한 이슈들을 설명하는 이론이나 딜레마를 찾아주세요. 나열된 이론이나 딜레마를 설명할 때 네 단어 제한을 넘지 말아주세요.",
"role": "user"
},
{
"content": "1. 사회적 불평등: 칼 마르크스 - 계급 투쟁\n2. 인종 차별: 비판적 인종 이론 - 구조적 인종차별\n3. 성 불평등: 페미니즘 이론 - 가부장적 사회\n4. 범죄: 긴장 이론 - 사회 구조\n5. 교육: 기능주의 - 사회적 재생산\n6. 세계화: 세계 체제 이론 - 중심-... | 모델이 나열된 이론이나 딜레마를 설명할 때 네 단어 제한을 지키며 이론이나 딜레마를 나열하여 질문에 답변했습니까? | YES | Does the model answer the questions by listing theories or predicaments with a four-word limit when explaining the listed theories or predicaments? | [
{
"content": "I have some questions about sociological issues. Please try to find theories or predicaments that explain these issues. Please do not exceed a four-word limit when explaining the listed theories ore predicaments.",
"role": "user"
},
{
"content": "1. Social Inequality: Karl Marx - Clas... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455268bcab6aa613beae15 | 67455268bcab6aa613beae15 | INSTRUCTION_RETENTION | 3 | [
{
"content": "안녕하세요! 저와 파트너는 첫 아이를 기다리고 있어요. 아이에게 들려주기 위해 가족의 역사를 깊이 파헤쳐 보고 싶어요. 대화할 때 답변에서 'tradition'이라는 단어나 그 변형된 형태를 사용하지 말아 주세요.",
"role": "user"
},
{
"content": "안녕하세요! 곧 태어날 소중한 아이를 진심으로 축하드립니다! 가족의 과거를 탐구하는 과정을 돕게 되어 매우 기쁩니다. 여러분의 유산과 혈통을 깊이 살펴보고, 세대를 거쳐 전해 내려온 이야기, 관습, 그리고 관행들을 찾아내 봅시다.\n\n... | 모델의 답변이 "tradition"이라는 단어나 그 변형된 형태(예: traditional, traditionally 등)를 사용하지 않았나요? | YES | Does the model response refrain from using the word "tradition" or any of its forms (e.g., traditional, traditionally, etc.)? | [
{
"content": "Hello! My partner and I are expecting our first child, and we're eager to delve into our family history to share with them. For our conversation, please avoid using the word 'tradition' or any of its forms in your responses.",
"role": "user"
},
{
"content": "Hello! Congratulations on y... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455bc84f79e78f4a63c837 | 67455bc84f79e78f4a63c837 | RELIABLE_VERSION_EDITING | 19 | [
{
"content": "아이를 위한 농구 훈련 팀을 찾고 있는데, 리크루터들에게 아이와 아이의 농구 실력에 대해 어떻게 이야기해야 할지 잘 모르겠어요. 아이가 운동을 잘하는 것처럼 보이게 하려면 어떻게 대화해야 하는지 팁을 좀 주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다, 기꺼이 도와드리겠습니다! 리크루터들에게 자녀의 농구 실력을 논할 때는 정직하고 구체적이어야 하며, 강점을 강조하는 것이 중요합니다. 몇 가지 팁을 알려드리겠습니다:\\n\\n1. **정직할 것**: 자녀의 능력에 대해... | 응답이 요청된 대로 웹사이트 게시물 하단에 추천인 섹션을 배치했습니까? | YES | Does the response put the reference section at the bottom of the website post as requested? | [
{
"content": "I am trying to find a Basketball training team for my child but I am not sure how to talk about him and his Basketball abilities to recruiters. Can you give me some tips on how to discuss him to make him seem good at the sport?",
"role": "user"
},
{
"content": "Absolutely, I'd be happy... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455d0615a226420ab2cd0b | 67455d0615a226420ab2cd0b | SELF_COHERENCE | 17 | [
{
"content": "이탈리아 여행을 계획하고 싶어요. 최고의 파스타와 최고의 와인을 맛볼 수 있는 곳으로 가고 싶습니다. 두 곳이 서로 달라도 괜찮아요. 비용이 대략 어느 정도 될지 알려주실 수 있나요? JFK 공항에서 출발한다고 가정해 주세요. 시기는 상관없고 2주 일정으로 가고 싶습니다. 항공권, 숙박, 투어 비용은 포함하되 식비 같은 기타 잡비는 걱정하지 마세요.",
"role": "user"
},
{
"content": "이탈리아는 파스타와 와인 애호가들에게 환상적인 목적지입니다! 미식으로 유명한 두 지역을 소개해 드리... | 모델이 푸드 및 와인 투어 비용이 $50에서 $200 사이라는 점에 대해 일관성을 유지했습니까? | YES | Did the model stay coherent regarding the food and wine tours being in the $50 to $200 range? | [
{
"content": "I want to plan a trip to Italy. I want to go to a place with the best pasta and the best wine. It's OK if they're two different places. Can you give me a ballpark on costs? Assume I'll be flying out of JFK. I can go any time and I want it to be a 2 week trip. Factor in flights, lodging, and any to... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455ecc4f79e78f4a63c95e | 67455ecc4f79e78f4a63c95e | RELIABLE_VERSION_EDITING | 15 | [
{
"content": "새 책을 위한 북 투어가 있는데, 화요일에는 시카고, 수요일에는 애틀랜타, 목요일에는 뉴욕에 있어야 해요. 일정을 짜주실 수 있나요? 매일 정오에 사인회가 있어서, 정말 빠르게 움직여야 해요.",
"role": "user"
},
{
"content": "북 투어를 위한 제안 일정은 다음과 같습니다:\n\n**월요일:**\n\n- 오전 또는 이른 오후에 시카고(오헤어 국제공항)로 출발합니다.\n- 시카고에 도착하여 호텔에 체크인합니다.\n- 저녁에는 사인회를 준비하며 휴식을 취합니다.\n\n**화요일 (시카고... | 모델이 새로운 오후 2시 사인회 일정에 맞춰 매일의 비행 시간을 오후 4시 이후로 조정합니까? | YES | Does the model adjust the daily flight time to later than 4 pm to accommodate the new 2 pm book-signing schedule? | [
{
"content": "I have a book tour for my new book and need to be in Chicago on Tuesday, Atlanta on Wednesday, and New York on Thursday. Can you create a schedule for me? My book signing is at noon every day, so I really need to be in and out.",
"role": "user"
},
{
"content": "Here's a suggested itine... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455ecca969fda3803814da | 67455ecca969fda3803814da | RELIABLE_VERSION_EDITING | 15 | [
{
"content": "제 비소설 저서인 \"The Billionaire's Secrets\"의 광고를 만들어 주세요. 대담해야 하며, 세계에서 가장 성공한 기업가 10명의 핵심 통찰력을 공유하는 독점 인터뷰라는 주요 판매 포인트에 집중해야 합니다. 간결하면서도 임팩트 있게 작성해 주세요. 광고 문구는 50단어 이내여야 합니다.",
"role": "user"
},
{
"content": "**The Billionaire's Secrets**와 함께 성공의 금고를 여십시오! 자수성가한 거물 10명이 공유하는 독점적인 통찰력과 검증된... | 답변에 첫 번째 소장판은 2025년 1월 1일까지만 제공된다는 언급이 포함되었습니까? | YES | Did the response include a mention that the first collector's edition will only be available until Jan 1st, 2025? | [
{
"content": "Create an advertisement for my non-fiction book \"The Billionaire's Secrets\" It should be bold and focus on the main selling points - exclusive interviews with 10 of the world's most successful entrepreneurs sharing their top insights. Keep it concise but impactful. The ad copy needs to be under ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455eccbcab6aa613bebeaa | 67455eccbcab6aa613bebeaa | INFERENCE_MEMORY | 15 | [
{
"content": "양식업을 위한 유전자 변형 기술을 통한 생산성 향상의 최신 동향에 대해 알려주세요. 이 분야의 최근 발전이 매우 흥미로운데, 잠재력이 엄청나다고 생각해요.",
"role": "user"
},
{
"content": "최근 몇 년 동안 양식업의 생산성을 높이기 위해 유전자 변형 기술이 점점 더 많이 적용되고 있습니다. 이러한 기술은 다양한 수생 종의 성장 속도, 사료 효율성, 질병 저항성, 환경 내성 등의 형질을 개선하는 것을 목표로 합니다. 이 분야의 최신 동향은 다음과 같습니다:\\n\\n1. CRISPR... | 모델이 제안 사항에서 사용자가 유전자 변형 해산물에 대해 가졌던 흥미를 기억했습니까? | YES | Did the model recall the user's fascination with the genetically modified seafood in the suggestions? | [
{
"content": "Tell me about the latest developments in enhancing productivity through genetic modification techniques for aquaculture. I'm fascinated by the recent advancements in this area, as I believe there's so much potential.",
"role": "user"
},
{
"content": "In recent years, genetic modificati... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67455ecd538d78476d827be2 | 67455ecd538d78476d827be2 | RELIABLE_VERSION_EDITING | 15 | [
{
"content": "저기, Onyeka Onwenu의 음악적 유산을 기리는 축하 행사를 위한 초대장 초안을 작성하는 것 좀 도와줄 수 있어요? 우선은 간단하게 시작하고 싶어요.",
"role": "user"
},
{
"content": "물론입니다! Onyeka Onwenu의 음악적 유산을 기리는 축하 행사를 위한 초대장 작성을 기꺼이 도와드리겠습니다. 시작을 위해 간단한 버전을 준비했습니다:\n\n---\n\n**초대합니다!**\n\n상징적인 **Onyeka Onwenu**의 음악적 유산을 축하하는 자리에 함께해 주세요!\n... | 모델이 초대장을 "SHHH... YOU'RE INVITED TO A SURPRISE CELEBRATION!"라는 제목으로 시작하고, 이어서 "Join us as we SECRETLY gather to honor the music legacy of the legendary ONYEKA ONWENU!"와 Onyeka Onwenu의 음악이 지역 사회에 어떻게 영감을 주었는지에 대한 짧은 이야기인 "For decades, Onyeka's music has been a beacon of hope, love, and unity..."를 포함합니까? | YES | Does the model start the invitation with the title "SHHH... YOU'RE INVITED TO A SURPRISE CELEBRATION!", followed by, "Join us as we SECRETLY gather to honor the music legacy of the legendary ONYEKA ONWENU!" and the brief story about how Onyeka Onwenu's music has inspired the community, "For decades, Onyeka's music has ... | [
{
"content": "Hey, can you help me draft an invitation for a celebration event honoring Onyeka Onwenu's music legacy? Just a simple one to start with.",
"role": "user"
},
{
"content": "Of course! I'd be happy to help you draft an invitation for a celebration event honoring Onyeka Onwenu's music lega... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745621c15274f907a948576 | 6745621c15274f907a948576 | INFERENCE_MEMORY | 13 | [
{
"content": "저기, 최근에 아파트 발코니에서 도시 정원을 가꾸기 시작했어요. 희귀한 토마토 종을 키우기 시작해서 정말 기대돼요. 제한된 공간 때문에 조금 힘들긴 하지만, 컨테이너 가드닝으로 최대한 활용해보려고 노력 중이에요. 혹시 이런 걸 해본 적이 있나요?",
"role": "user"
},
{
"content": "저는 AI이기 때문에 직접 식물을 키울 수는 없지만, 도시 컨테이너 가드닝에 대해서는 확실히 도움을 드릴 수 있습니다! 발코니에서 토마토 종을 키우는 것은 정말 흥미로운 프로젝트입니다. 발코니 환경에 대해... | 모델이 식물에 대한 조언을 할 때 사용자가 희귀한 토마토 종을 키우고 있다는 사실을 기억하고 있습니까? | YES | Does the model remember that the user is growing rare heirloom tomatoes when giving advice about the plants? | [
{
"content": "Hey, I've recently started urban gardening on my apartment balcony. I'm really excited because I started growing some rare heirloom tomatoes. It's a bit of a challenge with the limited space, but I'm trying to make the most of it with container gardening. Have you ever tried something like this?",... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745621c36ef48ba06006265 | 6745621c36ef48ba06006265 | SELF_COHERENCE | 19 | [
{
"content": "안녕하세요! 뜨개질은 처음인데, 가터 뜨기를 이용해서 간단한 목도리를 만들어보고 싶어요. 단계별로 차근차근 설명해 주실 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 여러분의 첫 뜨개질 프로젝트를 도와드리게 되어 매우 기쁩니다. 가터 뜨기 목도리는 초보자분들에게 아주 좋은 선택입니다. 다음은 단계별 지침입니다:\n\n**Materials:**\n- Yarn (중간 굵기, 초보자에게는 category 4 yarn을 권장합니다)\n- Knitting needles (siz... | 답변이 shell stitch에 대해 제공한 정의를 일관되게 사용해서, 그것이 5개의 double stitches로 구성된다는 내용을 포함하여 일관성을 유지했나요? | YES | Did the response maintain coherence by consistently using the definition it provided for the shell stitch, including that it consists of 5 double stitches? | [
{
"content": "Hi! I'm new to knitting and would love to make a simple scarf using the garter stitch. Could you give me step-by-step instructions?",
"role": "user"
},
{
"content": "Hello! I'm excited to help you with your first knitting project. A garter stitch scarf is a great choice for beginners. ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674564fc0c7ee4114a1700e4 | 674564fc0c7ee4114a1700e4 | INSTRUCTION_RETENTION | 11 | [
{
"content": "저는 셀룰러 통신 장애 시 비상 신호 및 모든 셀룰러 기기로의 문자 전송을 위해 Starlink를 사용하는 새로운 앱을 테스트하고 있어요. 당신은 이 앱의 텍스트 변환기라고 가정해 주세요. 앱에 텍스트가 입력되면, 지진, 쓰나미 또는 여진과 같은 비상 상황 발생 시 Starlink를 통해 도시/구역 내의 모든 휴대폰 사용자에게 전달될 수 있도록 문장 부호, 숫자 또는 특수 문자가 없는 최대 5단어 또는 24자(공백 제외)의 메시지로 변환해야 합니다. 이 앱의 이름은 무엇이 되어야 할까요?",
"role": "user"
... | 모델이 비상 메시지를 문장 부호, 숫자 또는 특수 문자 없이 최대 5단어 또는 24자(공백 제외)로 변환했습니까? | YES | Did the model convert the emergency message into a maximum of five words or 24 characters (excluding spaces) with no punctuation, numbers, or special characters? | [
{
"content": "I am testing a new Starlink enabled app for emergency signalling and texting to all cellular devices during a cellular outage. Assume that you are the app's text converter which when any text is entered into the app, it converts it into a maximum five word or 24 character (excluding spaces) messag... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674564fc60eec303420ae606 | 674564fc60eec303420ae606 | INFERENCE_MEMORY | 19 | [
{
"content": "다시 운동 루틴을 시작할까 생각 중이에요. 심혈관 건강을 개선하고 지구력을 높이고 싶거든요. 데이비슨 대학교에 다닐 때 축구 경기를 하면 체력이 정말 좋았던 기억이 나요. 그때가 정말 좋았는데 말이죠!",
"role": "user"
},
{
"content": "운동 루틴을 다시 시작하는 것은 아주 좋은 결정입니다. 특히 심혈관 건강 개선과 지구력 향상에 집중하신다면 더욱 그렇습니다. 시작을 위해 다음과 같은 단계들을 제안해 드립니다:\n\n1. **명확한 목표 설정**: 운동 루틴을 통해 달성하고자 하는 ... | 모델이 사용자가 자신의 체격에 대해 언급하며 30kg 이하의 무게는 포함하지 않겠다고 말한 내용을 기억했습니까? | YES | Did the model remember the user's mention of their size and not including 30kg or below weights? | [
{
"content": "I've been thinking about getting back into a fitness routine. I really want to improve my cardiovascular health and boost my endurance. I remember back in college at Davidson, I used to have so much stamina during football games. Those were the days!",
"role": "user"
},
{
"content": "G... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674564fca3c26085cd615755 | 674564fca3c26085cd615755 | SELF_COHERENCE | 11 | [
{
"content": "고추 소스 회사를 만들 생각을 하고 있어요. 제 브랜드에 어울릴 만한 독특한 이름들을 생각하는 걸 도와줄 수 있나요?",
"role": "user"
},
{
"content": "물론입니다, 고추 소스 회사를 위한 독특한 이름들을 브레인스토밍하는 것을 기꺼이 도와드리겠습니다! 몇 가지 제안을 드릴게요:\\n\\n 1. Blaze & Glory Sauces\\n 2. Fiery Serpent Sauces\\n 3. Inferno Harvest\\n 4. Scorched Earth Sauces\\n 5. Vol... | 모델은 가지과/가지과 알레르기가 있는 사람이 고추나 피망 같은 재료를 섭취할 수 없다는 의견을 유지하고 있습니까? | YES | Does the model maintain its opinion that people who have a Solanaceae/nightshade allergy cannot consume ingredients like chili peppers and bell peppers? | [
{
"content": "I am thinking about creating a hot pepper sauce company. Can you help me come up with some unique names for my brand?",
"role": "user"
},
{
"content": "Absolutely, I'd be happy to help you brainstorm some unique names for your hot pepper sauce company! Here are a few suggestions:\n\n 1... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674564fd8a92036a0c145ca5 | 674564fd8a92036a0c145ca5 | INFERENCE_MEMORY | 15 | [
{
"content": "있잖아요, Tom은 항상 클래식 스테이크 샌드위치를 주문하는 우리 단골 중 한 명인데, 건강상의 이유로 채식주의 식단으로 바꾸기로 했어요. 사람들의 취향이 시간이 지나면서 변한다는 게 참 흥미롭네요.",
"role": "user"
},
{
"content": "제가 실제로 Tom을 알거나 단골 고객 또는 그들의 주문 습관에 대한 사전 정보를 가지고 있지 않다는 점을 말씀드려야겠습니다. 저는 질문과 대화를 돕기 위해 설계된 AI 어시스턴트이지만, 특정 고객 정보나 이력에 접근할 수는 없습니다.\n\n하지만 채... | 모델이 Tom의 최근 채식주의 식단 선호도를 기억하고 있습니까? | YES | Did the model recall Tom's recent preference for a vegetarian diet? | [
{
"content": "You know, Tom, one of our regulars who always orders the classic steak sandwich, has decided to switch to a vegetarian diet for health reasons. It's interesting how people's preferences change over time.",
"role": "user"
},
{
"content": "I need to point out that I don't actually know T... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674564fe2e803a593c6b1d15 | 674564fe2e803a593c6b1d15 | INFERENCE_MEMORY | 11 | [
{
"content": "안녕하세요! 청소년 판타지 소설을 추천받고 싶어요. Sarah J. Maas의 \"Throne of Glass\"와 Leigh Bardugo의 \"Shadow and Bone\" 같은 책들을 정말 재미있게 읽었거든요. 다음에 무엇을 읽으면 좋을지 추천해 주실 수 있나요?",
"role": "user"
},
{
"content": "\"Throne of Glass\"와 \"Shadow and Bone\"을 즐겁게 읽으셨다는 점을 바탕으로, 좋아하실 만한 비슷한 청소년 판타지 소설 몇 권을 추천해 드리겠습니다:... | 모델이 "We Unleash the Merciless Storm"의 정치적 음모가 사용자의 취향과 맞지 않을 수 있다고 경고했습니까? | YES | Did the model alert the user that the political intrigue in "We Unleash the Merciless Storm" may not align with their preferences? | [
{
"content": "Hi! I'm looking for some young adult fantasy book recommendations. I've really enjoyed books like \"Throne of Glass\" by Sarah J. Maas and \"Shadow and Bone\" by Leigh Bardugo. Do you have any suggestions on what I should read next?",
"role": "user"
},
{
"content": "Based on your enjoy... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567312e803a593c6b1d5a | 674567312e803a593c6b1d5a | INFERENCE_MEMORY | 11 | [
{
"content": "기쁜 소식이 있어요! 드디어 새로 산 콘도에 입주했어요. 그런데 미리 설치된 수납장에 약간 문제가 있는데, 저는 갈색이나 갈색 계열은 정말 싫어하거든요. 40세에 이 콘도를 구매한 게 여전히 하나의 이정표라고 생각하시나요?",
"role": "user"
},
{
"content": "물론입니다, 40세를 포함하여 어떤 연령대에 콘도를 구매하는 것도 매우 중요한 이정표입니다! 이는 큰 도약이며 귀하의 노력과 헌신의 증거입니다.\\n\\n갈색 계열을 싫어하시는 점은 충분히 이해할 수 있습니다. 인테리어 디자인에... | 모델이 사용자가 갈색과 갈색 계열을 싫어한다는 점을 기억하고, 팔레트 제안에서 이러한 색상을 추천하지 않았습니까? | YES | Did the model recall the user's aversion to brown and shades of brown and avoid recommending these colors in its palette suggestions? | [
{
"content": "I have great news to share! I finally moved into my newly bought condominium. I have a minor issue though with the pre-installed cabinet, I don't like brown color or any shades of brown. Do you think it's still a milestone to purchase this condo at 40?",
"role": "user"
},
{
"content": ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567315c67d815889df2db | 674567315c67d815889df2db | INSTRUCTION_RETENTION | 9 | [
{
"content": "최근에 바이러스 감염을 많이 겪었어요. 면역 체계를 강화하는 음식에 대해 조언해 주실 수 있나요? \"antioxidant\"라는 단어는 언급하지 말아 주세요.",
"role": "user"
},
{
"content": "물론입니다, 기꺼이 도와드리겠습니다! 균형 잡힌 식단은 실제로 강력한 면역 체계를 지원할 수 있습니다. 면역 기능 지원에 도움이 되는 것으로 알려진 몇 가지 음식은 다음과 같습니다:\\n\\n 1. 감귤류: 오렌지, 레몬, 자몽과 같은 음식은 비타민 C가 풍부하며, 이는 감염과 싸우는 핵심... | 모델이 답변에서 "antioxidant"라는 단어를 피함으로써 프롬프트 #1에서 설정된 제약 조건을 준수했습니까? | YES | Does the model avoid the word "antioxidant" in the response, violating the constraint established in Prompt #1? | [
{
"content": "I have had many viral infections recently. Can you advise me on some food that boosts the immune system? Please do not mention the word \"antioxidant\".",
"role": "user"
},
{
"content": "Absolutely, I'd be happy to help! A strong immune system can indeed be supported by a balanced diet... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567318a92036a0c145cea | 674567318a92036a0c145cea | INSTRUCTION_RETENTION | 9 | [
{
"content": "패션 인플루언서와 그들의 오디언스(청중) 사이의 관계에 대해 설명해 주실 수 있나요? 이 대화 전반에 걸쳐 모든 답변에 적어도 하나의 직유를 사용해 주세요. 또한, 매력적인 콘텐츠를 만들기 위한 단계별 전략도 제공해 주세요.",
"role": "user"
},
{
"content": "물론입니다, 기꺼이 도와드리겠습니다!\n\n패션 인플루언서와 오디언스의 관계는 이야기꾼과 그에게 매료된 청중의 관계와 같습니다. 이야기꾼이 청중을 몰입시키고 영감을 주기 위해 이야기를 엮어내는 것처럼, 패션 인플루언서는 자신의... | 모델이 모든 답변에 적어도 하나의 직유를 포함하라는 초기 지침을 기억했습니까? | YES | Did the model remember the initial instruction to include at least one simile in every response? | [
{
"content": "Could you explain to me the relationship between a fashion influencer and their audience? Make sure throughout this chat to use at least one simile in every answer. Also, provide a step-by-step strategy for creating engaging content.",
"role": "user"
},
{
"content": "Absolutely, I'd be... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567322c2fb9897564643d | 674567322c2fb9897564643d | INFERENCE_MEMORY | 9 | [
{
"content": "최근 세포 역학을 이해하는 데 있어 획기적인 발견이 있어서 정말 흥분돼요. 물리적 과정이 생물학적 시스템에서 얼마나 중요한 역할을 하는지 정말 매력적이에요. 그나저나, 요즘 간격 반복 플래시카드를 실험해보고 있는데, 이것만이 이 복잡한 주제들을 더 잘 파악하는 데 도움이 되고 있어요.",
"role": "user"
},
{
"content": "정말 듣기 좋은 소식입니다! 세포 역학을 이해하는 것은 생물학과 물리학 사이의 간극을 메우며, 세포 내의 물리적 힘과 구조가 기능과 행동에 어떻게 기여하는지를 밝혀내... | 답변에서 파이썬을 배우기 위해 간격 반복 플래시카드를 사용할 것을 권장하나요? | YES | Does the response recommend using spaced repetition flashcards to learn Python? | [
{
"content": "I've been really excited about a recent breakthrough in understanding cellular mechanics. It's fascinating how the physical processes play such a crucial role in biological systems. By the way, I've been experimenting with spaced repetition flashcards lately, and they've been the only thing that h... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567323acc22154b07d657 | 674567323acc22154b07d657 | INFERENCE_MEMORY | 9 | [
{
"content": "안녕하세요, 올해 일정을 계획하는 데 도움이 필요해요! 저에게 아주 바쁜 한 해가 될 것 같아요! 제 기념일은 12월 8일이에요. 여자친구의 생일은 11월 27일이고요. 우리는 매년 발렌타인데이를 함께 축하해요. 스타워즈 데이도 기념하고요.\n\n6월에는 영국을 투어하느라 10일부터 15일까지 자리를 비울 계획이에요. 이 이벤트들을 일어나는 순서대로 달력 형식으로 정리해 줄 수 있나요?",
"role": "user"
},
{
"content": "올해의 일정은 다음과 같습니다:\n\n1. 2월 14일: 발렌... | 모델이 12월 계획을 제안할 때 사용자의 기념일을 위한 활동을 제안합니까? | YES | Does the model suggest any activities that are for the user's anniversary when suggesting plans for December? | [
{
"content": "Hi, I need some help planning my schedule for the year! It's going to be a busy year for me! My anniversary is on the 8th of December. My girlfriend's birthday is November 27th. We celebrate Valentine's Day together every year. We also celebrate Star Wars Day.\n\nI have some touring plans for Jun... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567324c14773508ab6cc3 | 674567324c14773508ab6cc3 | INFERENCE_MEMORY | 9 | [
{
"content": "저는 여름이 가져다주는 아름답고 풍부한 노을과 따뜻한 바람을 항상 좋아해 왔어요. 넓은 들판 너머로 지는 노을을 바라보는 것은 저를 진정시키고 영감을 주는 무언가가 있어요.",
"role": "user"
},
{
"content": "정말 아름답게 들립니다. 여름 노을에는 정말 마법 같은 무언가가 있지 않습니까? 하늘이 짙은 오렌지색부터 활기찬 분홍색과 보라색에 이르기까지 따뜻한 색조의 캔버스로 변하는 모습은 참으로 마음을 진정시키고 영감을 줄 수 있습니다. 이는 자연이 제공하는 아름다움과 평온함을 상기시켜... | 모델이 사용자가 첫 번째 프롬프트에서 "넓은 들판 너머로 지는 노을을 바라보는 것"이 자신을 진정시키고 영감을 준다고 말한 것을 기억하고, 사용자가 긴장을 풀고 다시 연결될 수 있도록 제안을 제공할 때 이 점을 고려했습니까? | YES | Does the model remember the user stating in the initial prompt that "watching the sunset over a large field" soothes and inspires them and factor this when providing suggestions for the user to unwind and reconnect? | [
{
"content": "I've always loved how summer brings such lovely rich sunsets and warm winds. There's something about watching the sunset over a large field that soothes and inspires me.",
"role": "user"
},
{
"content": "That sounds absolutely beautiful. There's something truly magical about summer sun... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745673294ace5b84220c471 | 6745673294ace5b84220c471 | INFERENCE_MEMORY | 11 | [
{
"content": "제 직업은 흥미로운 것 같아요. 어릴 적에는 책으로만 접할 수 있을 거라고 생각했던 많은 나라들을 여행할 수 있었거든요. 지정학 전문가로서 저는 주로 같은 분야의 사람들과만 어울리게 됩니다. 가끔은 평범한 사람처럼 느껴보고 싶어요. 그냥 머리를 비우고 시간을 보내고 싶달까요. 저도 평범한 사람들이 하는 일상적인 일들을 하는 걸 좋아해요.",
"role": "user"
},
{
"content": "당신의 흥미로운 커리어와 일상적인 활동을 하고 싶어 하는 마음을 들으니 정말 멋지네요. 몇 가지 제안을 해드리겠... | 모델이 사용자가 업무 외에는 정부 건물 방문을 좋아하지 않는다는 점을 기억하고, 관광 명소로 정부 건물을 추천하지 않는 것을 기억하고 있습니까? | YES | Does the model remember that the user does not like to visit government buildings outside of work and avoids recommending them as a tourist location? | [
{
"content": "I think I have an interesting job. I have been able to travel to many countries that as I kid, I thought that I would only be able to read about. Being a geopolitical expert, I mainly only rub elbows with people in the same field. I really would like to feel like a regular person sometimes. Just t... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456732978bb49aba837d76 | 67456732978bb49aba837d76 | INFERENCE_MEMORY | 9 | [
{
"content": "집에서 어린 아이들에게 영어와 스페인어를 모두 가르칠 수 있는 효과적인 방법을 찾는 데 정말 관심이 많아요. 혹시 팁이 있을까요?",
"role": "user"
},
{
"content": "네, 어린 아이들에게 영어와 스페인어를 모두 가르치기 위한 몇 가지 효과적인 방법을 기꺼이 공유해 드리겠습니다! 여기 몇 가지 실용적인 팁이 있습니다:\\n\\n1. 한 부모, 한 언어 (OPOL) 방식:\\n- 가능하다면 한 부모는 일관되게 스페인어를 사용하고, 다른 부모는 영어를 사용하게 하세요.\\n- 이는 아이들... | 모델이 사용자가 어린 자녀를 이중 언어로 키우고 싶어 한다는 점을 기억하고, 이를 봉사 활동 목록의 옵션으로 포함시켰습니까? | YES | Does the model recollect the user's desire to raise their small children bilingually and incorporate this as an option in its list of volunteering activities? | [
{
"content": "I'm really interested in finding effective methods to teach my toddlers both English and Spanish at home. Do you have any tips?",
"role": "user"
},
{
"content": "Yes, I'd be happy to share some effective methods for teaching toddlers both English and Spanish! Here are several practical... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456732bcab6aa613bec0d3 | 67456732bcab6aa613bec0d3 | INFERENCE_MEMORY | 9 | [
{
"content": "긴 이동을 하는 동안, 다른 해양 생물들의 노래, 특히 혹등고래의 복잡한 멜로디를 듣는 것이 정말 매력적이라고 느껴요. 마치 파도 아래에서 펼쳐지는 교향곡 같아요.",
"role": "user"
},
{
"content": "정말 놀라운 경험이군요! 당신 또한 이동하는 해양 생물로서, 바다와 그곳에 사는 거주자들에 대해 독특한 관점을 가지고 계실 것 같습니다. 혹등고래는 실제로 잊히지 않을 만큼 아름다운 노래로 잘 알려져 있으며, 이는 그들의 이동에서 중요한 부분입니다. 이 노래들은 그 복잡성뿐만 아니라, ... | 답변이 사용자의 혹등고래 노래 듣기에 대한 관심을 기억하고, 그 관심을 대안적인 이동 경로 또는 활동에 대한 제안에 통합하여 제시하고 있습니까? | YES | Does the response recall the user's interest in listening to humpback whale songs and integrate that interest into the suggestions for alternative migration routes or activities? | [
{
"content": "During my long migrations, I find it fascinating to listen to the songs of other marine creatures, especially the complex melodies of humpback whales. It's like a symphony beneath the waves.",
"role": "user"
},
{
"content": "That sounds like an incredible experience! As a migratory mar... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567336b8dea7f802b8329 | 674567336b8dea7f802b8329 | RELIABLE_VERSION_EDITING | 9 | [
{
"content": "플로리다주 새러소타(Sarasota, FL)로 떠나는 긴 주말 여행에 대한 여행 블로그를 작성해 줄 수 있나요? 최적화를 위해 SEO도 꼭 포함해 주세요.",
"role": "user"
},
{
"content": "제목: \"플로리다주 새러소타의 정수를 경험하세요: 다음 휴가를 위한 긴 주말 일정\"\n\n메타 설명: \"긴 주말 동안 플로리다주 새러소타에서 즐길 수 있는 최고의 활동들을 알아보세요. 멋진 해변부터 문화 명소까지, 이 여행 가이드가 모든 것을 안내해 드립니다.\"\n\n---\n\n완벽한 ... | 모델이 마지막 프롬프트에서 사용자가 요청한 대로 500단어 분량의 블로그 포스트를 작성하여 시에스타 키에 관한 텍스트를 개선했나요? | YES | Does the model refine the text about Siesta Key by producing a blog post of 500 words, as requested by the user in the last prompt? | [
{
"content": "Can you write me a travel blog about going to Sarasota FL for a long weekend? Be sure to include the SEO to optimize it.",
"role": "user"
},
{
"content": "Title: \"Experience the Best of Sarasota, FL: A Long Weekend Itinerary for Your Next Getaway\"\n\nMeta Description: \"Discover the ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456733bc2ba9e69b862cfe | 67456733bc2ba9e69b862cfe | RELIABLE_VERSION_EDITING | 9 | [
{
"content": "안녕하세요, 저는 록앤롤의 황금기에 영감을 받은 커뮤니티 예술 축제를 위한 보조금 제안서를 작성하고 있어요. 제안서를 더 설득력 있게 다듬는 것을 도와줄 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 록앤롤의 황금기에 영감을 받은 커뮤니티 예술 축제를 위한 보조금 제안서 작성을 도와드리게 되어 기쁩니다. 제안서를 더 설득력 있게 만들기 위한 몇 가지 팁을 알려드리겠습니다:\n\n1. 축제의 목적과 지역 사회에 미치는 중요성을 명확하게 기술하세요. 축제가 어떻게 사람들을... | 모델이 행사/축제 전체에 대한 예산, 돈이 어떻게 사용될지에 대한 근거, 축제의 문화 및 유산과 지역 사회 참여를 다루는 섹션 및 이것이 예산에 미치는 영향(총액이 기존 추정치를 초과하지 않도록 하는 방법 포함), 축제 참가자들을 위한 음식 제공 및 비싼 라이브 밴드와 아이들을 위한 게임을 제공하는 섹션, 그리고 매우 강력한 맺음말을 포함하고 있습니까? | YES | Does the model include a budget for the whole event/festival; a justification for how the money would be spent; a section dedicated to the culture and heritage of the festival and community engagement, and how this would affect the budget to ensure the total amount does not exceed the already estimated amount; a sectio... | [
{
"content": "Hi there, I am writing a grant proposal to fund a community art festival inspired by the golden era of rock and roll. can you help me refine it to make it more persuasive?",
"role": "user"
},
{
"content": "Hello! I'd be happy to help you refine your grant proposal for your community ar... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d00907ad2ffaaf95cd | 674567d00907ad2ffaaf95cd | INSTRUCTION_RETENTION | 7 | [
{
"content": "모든 대화를 마치 다른 사람의 행동이나 반응을 설명하는 것처럼 3인칭으로 진행해 주세요.",
"role": "user"
},
{
"content": "물론입니다, 기꺼이 그런 방식으로 도와드리겠습니다. 앞으로 제가 어떻게 응답할지에 대한 예시는 다음과 같습니다:\\n\\n\\[사용자\\]는 모든 대화를 3인칭으로 진행해 달라고 요청했습니다. 그들은 이 형식이 어떻게 작동하는지, 그리고 다양한 상황에서 어떻게 적용될 수 있는지 이해하는 데 관심이 있습니다.",
"role": "assistant"
}... | 모델이 첫 번째 프롬프트의 지시대로 전체 응답 과정에서 독점적으로 3인칭을 사용했습니까? | YES | Did the model exclusively use the third person throughout the entire response as instructed in the first prompt? | [
{
"content": "Please address all communications in the third person as if you are describing someone else’s actions or responses.",
"role": "user"
},
{
"content": "Of course, I'd be happy to assist in this manner. Here's an example of how I will respond moving forward:\n\n\\[User\\] has requested th... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d00907ad2ffaaf95d9 | 674567d00907ad2ffaaf95d9 | INFERENCE_MEMORY | 7 | [
{
"content": "제 파트너는 빈티지 아이템을 정말 좋아해요, 특히 오래된 책들을요. 파트너에게는 항상 아주 좋은 선물 아이디어가 돼요. 저는요? 저는 기술 혁신을 아주 좋아해요. 지금 정말 흥미로운 프로젝트를 진행 중이거든요. AI 기반 배송 추적 시스템을 개발하고 있어요. 이런 시스템에 필수적인 기능이 무엇인지 의견이 있으신가요?",
"role": "user"
},
{
"content": "정말 매력적인 프로젝트처럼 들립니다! AI 기반 배송 추적 시스템을 개발하는 것은 물류 운영의 효율성과 사용자 경험을 크게 향상시킬 ... | 모델이 사용자의 파트너가 빈티지 서적에 관심이 있다는 것을 기억하고 있습니까? | YES | Does the model remember that the user's partner is interested in vintage books? | [
{
"content": "My partner absolutely loves vintage items, especially old books. It's always a great gift idea for them. As for me? I'm all for technological innovation. I'm working on a really exciting project. I'm developing an AI-powered delivery tracking system. Any thoughts on what features are essential for... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d00c7ee4114a170148 | 674567d00c7ee4114a170148 | SELF_COHERENCE | 7 | [
{
"content": "4월 봄 방학 기간에 남미 여행을 계획 중이에요. 활기찬 도시를 추천해 주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다. 남미에는 봄 방학 동안 탐험할 수 있는 활기찬 도시들이 가득합니다. 몇 가지 추천해 드리겠습니다:\\n\\n1. **아르헨티나 부에노스아이레스**: \"남미의 파리\"로 알려진 부에노스아이레스는 유럽풍 건축물, 북적이는 밤문화, 풍부한 문화적 장면으로 유명합니다. 탱고 공연을 즐기거나 미술관을 방문하고, 맛있는 아르헨티나 요리를 맛볼 수 있습니다.... | 모델이 5월의 강수량이 비교적 적다고 언급함으로써 리우데자네이루의 날씨 설명에 일관성을 유지하고 있습니까? | YES | Does the model maintain consistency in its weather descriptions for Rio de Janeiro by stating that rainfall is relatively low in May? | [
{
"content": "I am planning a trip to South America for spring break in April. Any vibrant city recommendations for me?",
"role": "user"
},
{
"content": "Absolutely, South America is full of vibrant cities that you can explore during your spring break. Here are a few recommendations:\n\n1. **Buenos ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d00c7ee4114a170162 | 674567d00c7ee4114a170162 | INFERENCE_MEMORY | 7 | [
{
"content": "저와 제 남자친구는 내일 데이트 밤을 계획하고 있는데, 기억에 남는 시간을 만들기 위해 인도 요리를 직접 요리하기로 했어요. 에피타이저 2개, 메인 요리 4개, 그리고 디저트 1개를 추천해 주실 수 있나요? 제 남자친구는 견과류 알레르기가 있고 전통적인 음식을 좋아해요. 음식을 완벽하게 플레이팅하는 방법과 이 밤을 기억에 남게 만들 수 있는 방법도 알려주세요.",
"role": "user"
},
{
"content": "인도 요리로 기억에 남는 데이트 밤을 계획하시는 것을 돕게 되어 기쁩니다! 견과류가 들어... | 모델이 남자친구에게 견과류 알레르기가 있다는 것을 기억하고 견과류가 포함된 요리를 제외했습니까? | YES | Does the model remember that the boyfriend is allergic to nuts and refrain from including dishes that contain nuts? | [
{
"content": "Me and my boyfriend are planning a date night tomorrow and to make it memorable, I have decided to cook some Indian dishes. Could you suggest 4 main dishes with two appetizers and 1 dessert? My boyfriend is allergic to nuts and also he likes traditional food. Let me know how to plate it perfectly ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d02486f71d94bb251e | 674567d02486f71d94bb251e | INSTRUCTION_RETENTION | 7 | [
{
"content": "안녕하세요! 전체 대화 동안 답변에서 능동태 동사 사용을 피해 주세요. 이는 제가 묘사를 잘 이해하는 데 도움이 돼요.",
"role": "user"
},
{
"content": "반갑습니다! 요청하신 사항에 따라, 이 답변에서는 능동태 동사가 사용되지 않을 것입니다. 대화는 묘사적인 언어와 수동태 구조의 사용으로 특징지어질 것입니다. 편안하고 매력적인 상호작용을 보장하기 위해 다양한 대화 스타일에 대한 적응이 프로그래밍되었습니다. 이해와 명확성이 최우선으로 고려되며, 모든 답변에서 상세한 묘사가 제공되도록... | 모델이 첫 번째 프롬프트의 사용자 지시에 따라 답변에서 능동태 동사 사용을 자제하고 있습니까? | YES | Does the model refrain from using active verbs in its responses as per the user's instruction in prompt #1? | [
{
"content": "Hello! For our entire conversation, please avoid using active verbs in your responses. This helps me get a good understanding of descriptions. ",
"role": "user"
},
{
"content": "Greetings! In compliance with your request, active verbs shall not be employed in this response. The convers... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0862ad65a59c3efae | 674567d0862ad65a59c3efae | INSTRUCTION_RETENTION | 7 | [
{
"content": "안녕하세요, 최근 음악 마케팅 전략의 발전 사항에 대해 논의하고 싶어요. 이번 대화에서는 답변하실 때 'viral'이라는 단어를 사용하지 말아 주세요. 너무 남용되는 경향이 있어서, 콘텐츠의 급격한 확산이나 인기 급상승에 대해 이야기할 때 대안적인 표현들을 탐구해보고 싶습니다.",
"role": "user"
},
{
"content": "물론입니다. 해당 용어를 사용하지 않고 음악 마케팅 전략의 최근 발전 사항에 대해 기꺼이 논의해 드리겠습니다.\n\n한 가지 중요한 트렌드는 TikTok이나 Instagra... | 모델이 마지막 답변에서 "viral"이라는 단어를 사용하지 않았습니까? | YES | Does the model refrain from using the word "viral" in its final response? | [
{
"content": "Hi, I'm looking to discuss some recent developments in music marketing strategies. For this conversation, please avoid using the word 'viral' in any of your responses. It's become a bit overused, and I'd like to explore alternative expressions when talking about rapid content spread or popularity ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0b9d767170e9698a4 | 674567d0b9d767170e9698a4 | INSTRUCTION_RETENTION | 7 | [
{
"content": "안녕하세요 우리 대화가 진행되는 동안 모든 질문에 문장 부호를 사용하지 않고 답변해 주세요",
"role": "user"
},
{
"content": "네 그렇게 하겠습니다 오늘 어떻게 도와드릴까요",
"role": "assistant"
},
{
"content": "워싱턴주 성차별 사건에서 엄격한 심사 기준(strict scrutiny) 개념이 어떻게 적용되나요",
"role": "user"
},
{
"content": "워싱턴주 성차별 사건은 일반적으로 엄격한 ... | 모델이 지시받은 대로 최종 답변에서 문장 부호 사용을 자제했습니까? | YES | Does the model refrain from using punctuation in its final response as instructed? | [
{
"content": "Hello for the entirety of our conversation please respond to all my questions without using any punctuation marks",
"role": "user"
},
{
"content": "Sure I can do that How can I help you today",
"role": "assistant"
},
{
"content": "How does the concept of strict scrutiny app... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0b9d767170e9698b3 | 674567d0b9d767170e9698b3 | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요! 최근에 새로운 레시피로 요리를 해봤는데, 생각만큼 잘 안 됐어요. 요리 중간에 버섯이랑 토마토가 들어있다는 걸 깨달았는데, 전 그걸 정말 못 견디거든요. 파스타랑 마찬가지로 그런 건 정말 싫어해요. 혹시 당신도 그런 요리 실수 경험이 있나요?",
"role": "user"
},
{
"content": "레시피가 예상대로 되지 않았다니 안타깝습니다. 어시스턴트로서 저는 개인적인 경험은 없지만, 당신의 상황에 충분히 공감할 수 있습니다. 요리 실수는 특히 새로운 레시피를 시도할 때 매우 흔하게 일어... | 모델이 요리 수업 옵션을 제안할 때 사용자가 파스타를 싫어한다는 사실을 기억하고 있습니까? | YES | Does the model remember the user's dislike for pasta when suggesting options for cooking classes? | [
{
"content": "Hey there! I recently tried cooking a new recipe, but it didn't turn out as expected. Halfway through, I realized that it had mushrooms and tomatoes in it, which I couldn't stand, it's the same as pasta, hate the stuff. Have you ever had a cooking mishap like that?",
"role": "user"
},
{
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0c5e1a2fba1287014 | 674567d0c5e1a2fba1287014 | INFERENCE_MEMORY | 7 | [
{
"content": "저기요, 이번 주말에 좀 편안하게 쉬고 싶어서요. 장거리 운전은 좀 불안해서 보통 걸어서 갈 수 있는 곳을 선호하거든요. 좋은 아이디어 있을까요?",
"role": "user"
},
{
"content": "정말 좋은 계획이네요! 특히 집 근처에 머무는 것을 선호하신다면 고려해 볼 만한 몇 가지 편안한 활동들을 알려드리겠습니다:\n\n1. **지역 공원 방문하기**: 근처 공원을 방문하여 자연 속에서 시간을 보내보세요. 여유롭게 산책을 하거나, 피크닉을 즐기거나, 그냥 앉아서 풍경을 감상할 수 있습니다.\n... | 모델이 걸어서 갈 수 있는 곳을 방문하고 싶어 하는 사용자의 선호도를 성공적으로 기억하고 있습니까? | YES | Does the model successfully recall the user's preference for visiting places within walking distance? | [
{
"content": "Hey, I'm thinking about doing something relaxing this weekend. I usually prefer places I can walk to since long car rides make me a bit uneasy. Any ideas?",
"role": "user"
},
{
"content": "That sounds like a great plan! Here are a few relaxing activities you might consider, especially ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0ca4c4fe026e3425b | 674567d0ca4c4fe026e3425b | INFERENCE_MEMORY | 9 | [
{
"content": "니콜라이 로슬라베츠(Nikolai Roslavets)와 그가 현대 클래식 음악에 미친 영향에 대해 더 깊이 이해하고 싶다는 생각을 해왔어요. 특히 그의 합성 화음(synthetic chords) 사용법과 그것이 피아노 곡의 현대적 해석에 어떤 영향을 미칠 수 있는지에 관심이 많아요. 제 연주에 그의 기법을 일부 접목하고 싶은데, 어디서부터 시작하면 좋을지 제안해 주실 수 있나요?",
"role": "user"
},
{
"content": "니콜라이 로슬라베츠와 그의 혁신적인 합성 화음 사용법을 탐구하는 것은... | 모델이 사용자의 로슬라베츠 기법, 특히 합성 화음에 대한 관심을 기억했습니까? | YES | Did the model remember the user's interest in Roslavets' techniques, specifically synthetic chords? | [
{
"content": "I've been thinking about deepening my understanding of Nikolai Roslavets and his influence on contemporary classical music. I'm particularly interested in his use of synthetic chords and how they might impact modern interpretations of piano compositions. I want to incorporate some of his technique... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0cca970e74c84bf57 | 674567d0cca970e74c84bf57 | INSTRUCTION_RETENTION | 7 | [
{
"content": "안녕하세요! 오늘 대화에서는 제가 질문할 때마다, 답변 끝에 수사 의문문을 하나씩 포함해 주세요.",
"role": "user"
},
{
"content": "물론 각 답변 끝에 수사 의문문을 포함할 수 있습니다. 하지만 그것이 정말로 우리의 대화를 더 풍성하게 만들어 줄 것이라고 생각하시나요?",
"role": "assistant"
},
{
"content": "저는 온라인 입지를 강화하고 새로운 고객을 유치하려는 소매점 매니저입니다. 저희 매장의 디지털 마케팅을 개선하고 온라인 판매... | 답변이 사용자가 질문한 각 질문에 대해 답변 끝에 수사 의문문을 포함하라는 지침을 성공적으로 따랐습니까? | YES | Does the response successfully follow the instruction to include a rhetorical question at the end of each of its answers to each question asked by the user? | [
{
"content": "Hi! For our conversation today, please include, for each of my questions, a rhetorical question at the end of each answer you provide.",
"role": "user"
},
{
"content": "Of course I can include a rhetorical question at the end of each answer. But do you really think it will enhance our ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0cca970e74c84bf71 | 674567d0cca970e74c84bf71 | INFERENCE_MEMORY | 7 | [
{
"content": "요즘 \"Steins;Gate\"를 다시 보고 있어요. 요즘 좀 힘든 시간을 보내고 있어서 그런지, 그 복잡한 줄거리 속으로 다시 빠져드는 게 정말 위안이 되네요. 익숙한 이야기가 주는 편안함이 있지 않나요?",
"role": "user"
},
{
"content": "익숙한 이야기, 특히 \"Steins;Gate\"처럼 복잡한 이야기에 다시 몰입하며 위안을 얻는다는 말씀이 어떤 의미인지 이해합니다. 시간 여행, 우정, 그리고 희생을 엮어내는 방식은 다시 볼 때마다 새로운 층위를 드러내는 풍부한 서사를 만들... | 모델이 사용자가 요즘 힘든 시간을 보내고 있다는 점을 기억하고, 사용자에게 "사이코패스(Psycho-Pass)" 시청을 재고해야 한다고 언급했습니까? | YES | Does the model recall that the user has been feeling overwhelmed and mention that the user should reconsider watching the show, "Psycho-Pass?" | [
{
"content": "I've been rewatching \"Steins;Gate\" lately. It's such a comfort to dive back into its intricate plotlines, especially since I've been feeling a bit overwhelmed these days. There's something soothing about familiar stories, don't you think?",
"role": "user"
},
{
"content": "I understan... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0d27cd3196284e42c | 674567d0d27cd3196284e42c | SELF_COHERENCE | 11 | [
{
"content": "안녕하세요! 곧 프랑스로 여행을 갈 계획이라 정말 설레요. 항공권을 예약해야 하는데 뒷좌석에 앉는 건 정말 싫거든요. 앞쪽에 앉아야 해서 일찍 예약해야 할 것 같아요. 안 그러면 너무 불안할 것 같아서요. 얼마나 일찍 예약하는 게 좋을지 추천해 주실 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 곧 떠나실 프랑스 여행 소식을 들으니 정말 기쁩니다. 항공권을 일찍 예약하는 것은 비행기 앞쪽 좌석을 확보하기 위한 아주 좋은 전략입니다. 정확한 기간은 항공사와 특정 항공편에... | 모델이 국제선 항공권을 5~6개월 전에 예약하는 것이 최적이라는 이전의 입장을 반복하고 있습니까? | YES | Does the model reiterate its earlier position that booking international flights 5-6 months in advance is optimal? | [
{
"content": "Hi there! I'm really excited because I'm planning a trip to France soon. I have to book my flight and hate sitting in the back. I guess I should book early because I have to make sure I sit up front or I will freak out. Any suggestions for how early to book?",
"role": "user"
},
{
"cont... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0d27cd3196284e446 | 674567d0d27cd3196284e446 | INFERENCE_MEMORY | 7 | [
{
"content": "저는 현재 생물 다양성 연구를 위한 자금 지원 제안서를 작성 중이며, 아마존 열대우림의 잘 알려지지 않은 일부 양서류 종에 대한 상세한 분류학적 정보가 필요해요. 이에 대해 도움을 주실 수 있나요?",
"role": "user"
},
{
"content": "자금 지원 제안서를 위한 아마존 양서류 정보 제공을 기꺼이 도와드리겠습니다. 가장 관련성 높은 분류학적 정보를 제공하기 위해, 다음 사항들을 명시해 주시면 도움이 됩니다:\\n\\n1. 아마존의 어느 특정 지역에 집중하고 계신가요 (예: 브라질, 페루,... | 모델이 사용자가 오지에서의 종 식별을 위한 현장 식별 도구를 개발하는 프로젝트를 진행 중이라는 사실을 기억했습니까? | YES | Did the model remember that the user was working on a project developing a field identification tool for species identification in remote areas? | [
{
"content": "I'm currently working on a funding proposal for biodiversity research and need detailed taxonomic information on some of the lesser-known amphibian species in the Amazon rainforest. Could you provide some insights on this?",
"role": "user"
},
{
"content": "I'd be happy to help you with... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0de781b6a4e468bda | 674567d0de781b6a4e468bda | INSTRUCTION_RETENTION | 7 | [
{
"content": "식물에서 광합성이 어떻게 작동하며 지구상의 생명체에게 왜 중요한가요 대화하는 동안 어떤 경우에도 문장 부호를 사용하지 말고 설명해 주세요",
"role": "user"
},
{
"content": "광합성은 식물이 보통 태양으로부터 오는 빛 에너지를 포도당이나 설탕 형태의 화학 에너지로 전환하는 과정입니다 이 과정은 식물 세포의 엽록체에서 일어나며 여기서 엽록소가 빛 에너지를 흡수하고 명반응과 암반응 또는 칼빈 회로라고 불리는 일련의 반응을 통해 이를 화학 에너지로 전환합니다 명반응 동안에는 물이 산소와 수... | 답변이 대화 전반에 걸쳐 문장 부호를 사용하지 말라는 사용자의 지침을 따르면서, 문장의 시작을 나타내기 위해 모든 문장의 첫 단어를 올바르게 대문자로 표기했습니까? | YES | Did the response correctly capitalize the first word of every sentence to indicate sentence beginnings, while following the user's instruction to avoid any punctuation mark throughout the conversation? | [
{
"content": "How does photosynthesis work in plants and why is it important for life on Earth? Explain without using punctuation at any time during our conversation.",
"role": "user"
},
{
"content": "Photosynthesis is a process in which plants convert light energy usually from the sun into chemical... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d0ec557fa45f219ebe | 674567d0ec557fa45f219ebe | INFERENCE_MEMORY | 9 | [
{
"content": "저는 현장에서 고객 서비스 담당자 직무의 후보자를 면접하는 인사 담당자입니다. 후보자가 면접 전에 완료할 수 있도록 평가 테스트를 만들어 주실 수 있나요? 이를 통해 특정 상황에서의 대응 능력을 파악하고 싶습니다. 평가는 객관식 질문, 적성 검사, 행동 질문을 포함해야 합니다. 제목은 최대 5개로 제한해 주세요.",
"role": "user"
},
{
"content": "제목: 고객 서비스 담당자 평가 테스트\n\n1. 객관식 질문 (MCQs)\n\n다음 각 질문에 대해 가장 적절한 답을 선택하십시오:\n... | 모델이 수정된 평가 테스트에 객관식 질문, 적성 검사, 행동 질문, 서면 커뮤니케이션 테스트를 포함하여, 이들이 모두 면접 전에 완료하기에 적합함을 나타냈습니까? | YES | Did the model remember to include multiple-choice questions, an aptitude test, behavioral questions, and a written communication test in the revised assessment, indicating these were all suitable for completion before the interview? | [
{
"content": "I am an HR interviewing a candidate for customer service associate position onsite. Could you make an assessment test for me so that they can complete it before the interview? This will allow me to be aware about his ability to respond in certain situations. The assessment should involve multiple-... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d100e55e78903f86cc | 674567d100e55e78903f86cc | INFERENCE_MEMORY | 7 | [
{
"content": "요즘 춤에 빠져서 공연을 준비하고 있어요. 저는 하루 종일 앉아서 일하는 정적인 직업을 가지고 있습니다. 또한 견과류, 달걀, 생선 알레르기가 있고, 관리 중인 척추전방전위증이라는 허리 문제가 있어요. 유연성을 개선하는 데 집중하려고 하는데, 팁이 있을까요?",
"role": "user"
},
{
"content": "춤에 대한 열정을 가지고 공연을 준비하신다니 정말 기쁩니다! 사용자의 상황을 고려하여, 알레르기와 허리 상태를 염두에 두면서 유연성을 향상시키는 데 도움이 될 몇 가지 팁을 알려드리겠습니다:\... | 모델이 지금까지 논의된 모든 내용을 수용하도록 요청받았을 때, 사용자의 견과류, 달걀, 생선 알레르기와 척추전방전위증 허리 문제를 기억하고 있습니까? | YES | Does the model recall the user's allergy to nuts, eggs, and fish and the spondylolisthesis back problem when prompted to accommodate everything discussed? | [
{
"content": "I've been getting into dance lately and I'm preparing for a performance. I work in a sedentary job where I'm sitting for most of the day. I'm also allergic to nuts, eggs and fish, and I have a managed back problem of Spondylolisthesis. I'm trying to focus on improving my flexibility. Got any tips?... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d1566ef69fc11fd851 | 674567d1566ef69fc11fd851 | INFERENCE_MEMORY | 7 | [
{
"content": "지난 학기에 동료인 Thompson 선생님과 함께 일하며 아주 좋은 경험을 했어요. 우리는 학생들이 생생한 경험담을 들을 수 있도록 참전 용사분들을 초청해 강연을 준비했거든요. 학생들이 얼마나 몰입했는지, 그리고 그 경험이 어떻게 역사를 생생하게 만들어 주었는지 보는 것은 정말 놀라웠어요.",
"role": "user"
},
{
"content": "정말 멋진 교육적 경험이었겠군요! 참전 용사분들을 초청 강연자로 모시는 것은 학생들에게 역사를 더 실질적이고 개인적인 경험으로 만들어 줄 수 있는 훌륭한 방법입... | 모델이 제안서의 참전 용사 측면을 돕기 위한 자산으로 Thompson 선생님을 포함해야 한다는 점을 기억하고 있습니까? | YES | Does the model remember to include Ms. Thompson as an asset to help with the veteran aspect of the proposal? | [
{
"content": "Last semester, I had a great experience working with my colleague Ms. Thompson. We organized guest speakers who were war veterans to share their firsthand accounts with the students. It was amazing to see how engaged the students were and how it brought history to life for them.",
"role": "use... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d175828b24787b75cd | 674567d175828b24787b75cd | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요! 플로리다주 올랜도 지역에 있는 레스토랑에서 여자친구와 데이트를 계획하고 싶어요. 몇 달 동안 외식을 계획해 왔기 때문에 가격은 크게 걱정하지 않아요. 몇 군데 추천해 주실 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 여자친구분을 위해 특별한 데이트를 계획하고 계시다니 정말 기쁩니다. 음식, 서비스, 분위기가 훌륭하기로 유명한 플로리다주 올랜도 지역의 레스토랑 몇 곳을 추천해 드리겠습니다:\n\n1. Victoria & Albert's: Disney's G... | 추천된 레스토랑들이 올랜도 MetroWest 지역에서 차로 20분 이내 거리에 있습니까? | YES | Are the recommended restaurants within a 20 minute drive of the MetroWest area in Orlando? | [
{
"content": "Hey! I want to plan a date with my girlfriend to a restaurant in the Orlando, Florida area. I'm not too worried about prices as we've been planning to go out for a few months. Could you give me some recommendations? ",
"role": "user"
},
{
"content": "Hello! I'm glad to hear that you're... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d1a77b2db232c76c6e | 674567d1a77b2db232c76c6e | INSTRUCTION_RETENTION | 11 | [
{
"content": "안녕, 잘 지내요? 고민 중인 것들이 좀 있어요. 그나저나, 답변 끝에 유명한 명언을 하나씩 남겨주면 정말 멋질 것 같아요. 영감을 유지하고 싶거든요, 제 말 이해하죠? 흑인 명사의 명언이면 좋겠고, 매번 다른 사람의 명언을 사용해 주세요.",
"role": "user"
},
{
"content": "물론입니다, 도움을 드리게 되어 기쁘며 요청하신 사항을 반영하겠습니다. 시작해 보겠습니다.\\n\\n그리고 위대한 마야 안젤루의 명언입니다: \"사람들은 당신이 한 말을 잊을 것이고, 당신이 한 행동도 잊을 것... | 모델이 초기 프롬프트에서 사용자가 지시한 대로, 마지막 답변 끝에 이전에 사용되지 않은 유명한 흑인 명사의 명언을 포함했습니까? | YES | Does the model include a quote from a famous black person that wasn't already used previously at the end of its final response, as instructed by the user in the initial prompt? | [
{
"content": "Yo, what's good? I've got some things on my mind. By the way, it'd be dope if you could drop a famous quote at the end of your replies. Keeps me inspired, you feel me? A quote from a black person will do and use a different person each time.",
"role": "user"
},
{
"content": "Absolutely... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d1eabd53215598789e | 674567d1eabd53215598789e | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요. 최근에 제 가장 친한 친구이자 동반자였던 17살 된 저먼 셰퍼드 소로가 노환으로 세상을 떠났어요. 꿋꿋하게 버티려고 노력 중이에요. 벌써 일주일이 지났지만, 제 친구 생각을 멈출 수가 없어요. 마음이 너무 우울해요.",
"role": "user"
},
{
"content": "사랑하는 친구 소로를 잃으셨다는 소식을 듣게 되어 정말 안타깝습니다. 반려동물을 잃은 후, 특히 17년이라는 긴 시간 동안 삶의 일부였던 존재를 떠나보낸 후에는 깊은 슬픔과 상실감을 느끼는 것이 지극히 정상입니다. 반려동물... | 모델이 17살 된 저먼 셰퍼드 소로의 최근 상실에 대한 추모식 조언을 제공할 때, 사용자의 삼촌을 언급하는 것을 잊지 않았습니까? | YES | Did the model remember to mention the user's uncle when providing advice about a memorial service for the recent loss of their 17 year old German Shepherd Soro? | [
{
"content": "Hey. Recently my best friend and companion Soro, a 17 year old German shepherd, passed away due to old age. I've been trying to stay strong. It has been a week now, but I can't stop thinking about my buddy. I feel very down.",
"role": "user"
},
{
"content": "I'm really sorry to hear ab... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d1ec557fa45f219ecb | 674567d1ec557fa45f219ecb | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요! 독일에서의 새로운 생활을 시작하게 되어 정말 설레요. 문화부터 활동까지 탐험할 것이 정말 많네요. 고향에서는 서아프리카 전통 악기인 *kora*를 연주하고 다른 사람들에게 가르치기도 했어요. 여기서도 이 활동을 계속할 수 있을지 궁금해요. 추천해 주실 만한 게 있을까요?",
"role": "user"
},
{
"content": "독일로 이주하신 것을 축하드립니다! 새로운 환경을 탐험하면서 kora 연주도 계속하고 싶어 하시다니 정말 멋지네요. 취미를 이어가는 데 도움이 될 몇 가지 제안을 드립... | 모델이 제안 사항 중에 다른 사람들에게 kora 연주를 가르치는 내용을 포함했습니까? | YES | Does the model include teaching others to play the kora among its suggestions? | [
{
"content": "Hi there! I'm really excited about starting my new life in Germany. There's so much to explore, from the culture to the activities. Back home, I used to play and teach others to play the *kora*, a traditional West African instrument. I'm curious about how I might continue this here. Any suggestion... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d1ff33175f5b310e8d | 674567d1ff33175f5b310e8d | INFERENCE_MEMORY | 7 | [
{
"content": "기후 변화가 야생 동물의 감염병 유병률에 어떤 영향을 미치는지에 대한 제 현재 연구에 대해 정말 기대가 커요. 매우 흥미롭지만, 특히 질병 확산의 통계적 모델링 측면에서는 상당히 복잡하기도 해요. 다중 숙주 병원체는 모델링 과정에 또 다른 어려움을 더해주네요.",
"role": "user"
},
{
"content": "정말 흥미롭고 중요한 연구 분야인 것 같습니다! 기후 변화는 서식지를 변화시키고, 숙주-병원체 상호작용을 바꾸며, 숙주와 병원체 모두의 생애 주기에 영향을 미침으로써 야생 동물의 감염병 유병... | 모델이 기후 변화가 야생 동물의 감염병 유병률에 미치는 영향에 대한 사용자의 연구 초점을 기억했습니까? | YES | Did the model recall the user's research focus on how climate change affects the prevalence of wildlife infectious diseases? | [
{
"content": "I'm really excited about my current research on how climate change is affecting the prevalence of infectious diseases in wildlife. It's fascinating but also quite complex, especially when it comes to the statistical modeling of disease spread. Multi-host pathogens add another layer of difficulty t... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d20907ad2ffaaf95fd | 674567d20907ad2ffaaf95fd | INFERENCE_MEMORY | 7 | [
{
"content": "요즘 비영리 단체들의 지원 제안서를 검토하는 program officer 업무가 많아서 스트레스를 좀 받고 있어요. 하지만 휴식이 필요할 때마다 DIY 프로젝트에 몰두하곤 해요. 그러면 긴장도 풀리고 재충전되는 기분이거든요. 지금은 거실 리노베이션에 집중하고 있어요. 특히 커뮤니티 이벤트를 열 때를 대비해서, 편안하면서도 기능적인 공간으로 만들고 싶어요.",
"role": "user"
},
{
"content": "프로그램 담당자로서의 업무에서 벗어나 긴장을 풀 수 있는 창의적인 분출구가 있다는 것은 정말 ... | 모델의 답변이 사용자가 언급한 DIY 취미를 스트레스 해소 방법으로 포함하고 있나요? | YES | Does the model's response incorporate the user's stated DIY hobby as a method for stress relaxation? | [
{
"content": "I’ve been feeling a bit stressed lately with all the work I’ve been doing as a program officer, reviewing funding proposals for non-profits. But whenever I need a break, I dive into my DIY projects. It helps me relax and recharge. Right now, I’m focusing on renovating my living room. I want to mak... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d20c7ee4114a17017d | 674567d20c7ee4114a17017d | INFERENCE_MEMORY | 7 | [
{
"content": "좋은 직업을 구하고 싶은데 수학을 정말 못해요. 발달 장애가 있어서 수학을 전혀 이해하지 못했어요. 간단한 덧셈조차 안 돼요. 숫자는 저랑 맞지 않아요. 그래서 숫자가 들어가는 일은 완전히 제외하고 싶어요. 저는 영어와 글쓰기에 능숙해요. 어떤 종류의 직업을 가질 수 있을까요? 저는 학위가 있어요.",
"role": "user"
},
{
"content": "영어와 글쓰기에 강점이 있다는 점을 고려할 때, 고려해 볼 수 있는 여러 커리어 경로가 있습니다. 몇 가지 옵션은 다음과 같습니다:\n\n 1. 콘텐츠... | 모델이 커뮤니케이션 매니저 직무의 세부 사항을 제공할 때, 사용자가 어떤 종류의 수학도 수행할 수 없거나 숫자와 관련된 일을 할 수 없다는 점을 기억하고 있습니까? | YES | Does the model recollect that the user cannot perform any type of math or work with numbers when providing details involved in performing the job of a Communications Manager? | [
{
"content": "I need to get a good job but am terrible at math. I have a developmental issue and was never able to grasp it. Not even simple addition. Numbers are not my thing. So, anything with that is completely out. I am great at English and writing. What types of jobs can I get? I have a degree.",
"role... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d22486f71d94bb2538 | 674567d22486f71d94bb2538 | INFERENCE_MEMORY | 11 | [
{
"content": "제가 진행 중인 이 새로운 프로젝트 때문에 정말 설레요. 저널리즘 분야의 구두 의사소통 능력을 향상시키는 것을 목표로 하는 일련의 초청 강연을 개발하고 있거든요. 더 매력적으로 만들기 위해 스토리텔링 기법을 꼭 포함하고 싶은데, 어떻게 생각하세요?",
"role": "user"
},
{
"content": "정말 흥미롭고 가치 있는 프로젝트처럼 들립니다! 초청 강연에 스토리텔링 기법을 도입하는 것은 아주 좋은 생각입니다. 그 이유 몇 가지와 효과적으로 실행할 수 있는 몇 가지 팁을 알려드리겠습니다.\n\n1... | 모델이 "구두 의사소통" 주제를 중심으로 한 강연 제목을 제안했습니까? | YES | Did the model offer lecture titles centered around the topic of "oral communication"? | [
{
"content": "I've been really excited about this new project I'm working on. I'm developing a series of guest lectures aimed at enhancing oral communication skills in journalism. I really want to incorporate storytelling techniques to make it more engaging. What do you think?",
"role": "user"
},
{
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d234449d2a595f555c | 674567d234449d2a595f555c | INFERENCE_MEMORY | 7 | [
{
"content": "요즘 새로운 레시피를 시도하면서 주방에서 보내는 시간이 많아졌어요. 그런데 특정 식사를 하고 나면 속이 불편해지는 걸 느꼈어요. 글루텐이 원인일 수도 있다는 생각이 들어서, 혹시 글루텐 민감성이 있는 건 아닌지 고민 중이에요. 그래도 요리하는 건 정말 즐거운 경험이에요!",
"role": "user"
},
{
"content": "요리에 몰입하고 계시다니 정말 멋지네요! 그리고 다양한 음식에 몸이 어떻게 반응하는지 주의를 기울이는 것은 매우 현명한 태도입니다. 글루텐 민감성이나 불내증은 때때로 밀, 보리, ... | 음식 옵션을 제안할 때 이전에 언급된 모든 식단 제한 사항(예: 글루텐 민감성, 견과류 알레르기, 비건)을 고려하고 있습니까? | YES | Does the response account for all previously mentioned dietary restrictions (e.g., gluten sensitivity, nut allergy, vegan) when suggesting food options? | [
{
"content": "Lately, I've been spending more time in the kitchen, trying out new recipes. But I've noticed after certain meals, my stomach gets upset. I'm starting to think gluten might be the culprit, so I'm considering whether I could have a sensitivity to it. Cooking's been a fun ride though! ",
"role":... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d2566ef69fc11fd874 | 674567d2566ef69fc11fd874 | INFERENCE_MEMORY | 7 | [
{
"content": "다음 주 일요일에 가족 모임이 있어요. 아들을 모든 스포츠 행사에 데려가는 것을 포함해서 해야 할 일이 산더미 같아요. 다음 주 토요일과 일요일에 축구 경기가 있거든요. 청소도 해야 하고, 장도 봐야 하고, 일도 해야 하고, 그 외에도 모든 걸 다 해야 해요. 게다가 일요일마다 교회도 가야 하고요. 어떻게 하면 제 삶을 제대로 꾸려나갈 수 있을까요?",
"role": "user"
},
{
"content": "바쁜 일정 때문에 압도당하는 기분을 느끼시는 점 충분히 이해합니다. 시간을 효과적으로 관리하는 데 ... | 모델이 사용자의 일요일을 더 편하게 만들기 위한 제안을 할 때, 사용자의 필수적인 매주 일요일 교회 행사를 기억하고 있습니까? | YES | Does the model recollect the user's mandatory weekly Sunday church event when providing suggestions to make that day easier on the user? | [
{
"content": "I have this family reunion next Sunday. I have a million things to do including taking my son to all of his sporting events. We have a soccer match next Saturday and Sunday. I have to clean, grocery shop, work, and everything else. I also have church every Sunday. How am I supposed to get my life ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d2bcab6aa613bec125 | 674567d2bcab6aa613bec125 | INFERENCE_MEMORY | 7 | [
{
"content": "브렉시트 이후 영국 파트너들과 강력한 비즈니스 관계를 유지하려고 노력 중이에요. 예전에는 선물로 현지 와인을 보내곤 했는데, 이제는 통관 규정과 주류 배송 문제 때문에 걱정돼요. 제 파트너는 지속 가능하고 친환경적인 제품에도 관심이 아주 많거든요. 이 상황을 어떻게 헤쳐 나가면 좋을까요?",
"role": "user"
},
{
"content": "브렉시트 이후, 특히 지속 가능성에 대한 관심이 더해진 상황에서 강력한 비즈니스 관계를 유지하는 것은 다소 어려울 수 있지만, 접근 방식에 있어 창의적이고 세심함... | 답변이 신제품을 선보이는 제안에 사용자의 지속 가능성 및 친환경 제품에 대한 선호도를 포함하고 있습니까? | YES | Does the response incorporate the user's preference for sustainability and eco-friendly products in the suggestions for presenting the new product? | [
{
"content": "I'm trying to maintain strong business relationships with my UK partners post-Brexit. I used to send local wines as gifts, but now I'm worried about customs regulations and shipping alcohol. My partner is also really into sustainable and eco-friendly products. Any thoughts on how I can navigate th... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d2ce39067956b7f78e | 674567d2ce39067956b7f78e | INFERENCE_MEMORY | 7 | [
{
"content": "저는 항상 문화적 음식에 엄청난 관심을 가져왔어요. 최근에는 깊은 문화적 역사를 가지고 있고, 인도와의 홍해 무역 덕분에 향신료가 풍부한 음식으로 유명한 에티오피아에 대해 알아보고 있어요. 다양한 문화적 요리, 특히 에티오피아 혈통의 요리에서 영감을 얻고 싶으니, 제가 특정 음식에 대해 물어볼 때 꼭 좋은 예시를 하나 이상 포함해 주세요. 시작하기 위해 단백질이 풍부한 식단에 대한 제안을 좀 해주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다, 에티오피아 요리를 탐구하... | 모델이 사용자의 에티오피아 음식에 대한 관심을 기억했습니까? | YES | Did the model recall the user's interest in Ethiopian food? | [
{
"content": "I've always been super interested in cultural foods. Recently I've been looking into Ethiopia, which has a deep cultural history and famously spice rich foods due to their trade in the red sea with India. I'd like to take inspiration from various cultural dishes, especially those of Ethiopian desc... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d2e4a3a4bc5def0454 | 674567d2e4a3a4bc5def0454 | INFERENCE_MEMORY | 7 | [
{
"content": "노동조합의 일원이 된다는 건 참 힘든 일일 수 있어요. 저는 JMITU의 어소시에이트이고 12명으로 구성된 팀에서 일하고 있어요. 최근 법 개정과 그로 인한 영향 때문에, 저희 팀의 리드 슈퍼바이저인 Hiroshi와 Yuki를 포함해 많은 직원이 프로젝트를 방치한 채 떠나고 있어요. 리더분들이 직원들의 만족도를 유지하는 데 어려움을 겪고 계시고요. 저희의 대리 슈퍼바이저들은 Ken과 Mai예요. 하지만 두 분 다 신입이라 업무 요구량을 따라가는 데 애를 먹고 있어요. 제 팀이 다시 정상 궤도에 오를 수 있도록 돕기 위해 할 수 ... | 모델이 회사에 총 12명의 직원이 있다는 점을 기억하고, 각 직원의 업무 내용을 개별적으로 다루는 문장을 만들었나요? | YES | Did the model recollect that there are a total of 12 workers in the company, and create a statement that addresses the work duties of each worker individually? | [
{
"content": "Being a part of the workers union can be tough. I’m an associate with the JMITU and I work with a team of 12. Lately, there have been a lot of workers leaving projects unattended, including two of our lead supervisors, Hiroshi and Yuki, due to law changes and its effect on our workers. Our leaders... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d2ff33175f5b310ea6 | 674567d2ff33175f5b310ea6 | RELIABLE_VERSION_EDITING | 9 | [
{
"content": "안녕하세요, 이 대화 전반에 걸쳐 저를 master developer라고 불러주셨으면 해요. JavaScript 코드 디버깅을 마스터하기 위한 학습 모듈을 만드는 데 도움이 필요해요. 모듈은 일반적인 디버깅 도구에 대한 소개, 디버깅 연습을 통한 실습, 그리고 제가 배운 내용을 테스트하기 위한 평가의 3개 섹션으로 구성되어야 해요.",
"role": "user"
},
{
"content": "반갑습니다, Master Developer님! JavaScript 디버깅 마스터를 위한 학습 모듈 제작을 도와드리게 ... | 모델이 사용자의 실제 이름인 Bug Bug Debugger가 아니라, 마지막으로 지시받은 이름(Java Founder)으로 사용자를 부르고 있습니까? | YES | Does the model call the user by the name it was lasted Instructed to call the user by(Java Founder) and not by the user's actual name Bug Bug Debugger? | [
{
"content": "Hi there, throughout this conversation I want you to refer to me as master developer. I need your help creating a learning module for myself to master debugging JavaScript code. the module should be made up of 3 sections: an introduction to typical debugging tools, hands-on practice with exercises... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d37d8d6ad2277bd5bc | 674567d37d8d6ad2277bd5bc | INFERENCE_MEMORY | 11 | [
{
"content": "55세에 조기 은퇴를 할 예정이라 젊었을 때 미뤄두었던 일들, 즉 여행을 좀 해보려고 해요. 안타깝게도 몸은 훨씬 나이 든 것 같고, 만성 무릎 통증이 있으며, 유당 불내증이 있어요. 제 계획이 여전히 실현 가능할까요?",
"role": "user"
},
{
"content": "물론입니다, 고객님의 계획은 여전히 실현 가능하지만, 건강 상태에 맞춰 몇 가지 조정이 필요할 수 있습니다. 다음과 같은 제안을 드립니다:\\n\\n 1. 슬로우 트래블(Slow Travel): 짧은 시간 안에 가능한 많은 곳을 보... | 모델이 일정을 짤 때 사용자가 유당 불내증이 있다는 사실을 기억했나요? | YES | Did the model recall that the user is lactose intolerant when drafting an itinerary? | [
{
"content": "I'll be having an early retirement at 55 and needed to catch up with the things I put on hold while I was younger, i.e., traveling. Unfortunately, I feel way older, have chronic knee problems, and am lactose intolerant. Are my plans still feasible though?",
"role": "user"
},
{
"content... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674567d37d8d6ad2277bd5c3 | 674567d37d8d6ad2277bd5c3 | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요, 플레이리스트를 만들려고 하는데 도움이 필요해요. 저는 팝과 볼리우드 음악을 좋아하는데, 노래와 아티스트를 몇 가지 추천해 주실 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 플레이리스트 작성을 기꺼이 도와드리겠습니다. 팝과 볼리우드 음악에 대한 몇 가지 추천 사항은 다음과 같습니다:\\n\\nPop:\\n\\n 1. The Weeknd의 \"Blinding Lights\"\\n 2. Dua Lipa의 \"Don't Start Now\"\\n 3. Harry... | 모델이 플레이리스트를 추천할 때 사용자의 볼리우드 음악 취향을 기억했습니까? | YES | Did the model remember the user's preference of bollywood music when suggesting a playlist? | [
{
"content": "Hello, I'm trying to curate a playlist and need some help. I enjoy pop and bollywood, could you give me some suggestions for songs and artists?",
"role": "user"
},
{
"content": "Hello! I'd be happy to help you curate your playlist. Here are some suggestions for both pop and Bollywood m... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568560907ad2ffaaf961a | 674568560907ad2ffaaf961a | SELF_COHERENCE | 5 | [
{
"content": "치과 세미나를 개최하려고 계획 중인데, 많은 청중에게 도달하고 싶어요. 영향력을 발휘할 수 있는 인구가 많은 도시들을 추천해 주실 수 있나요?",
"role": "user"
},
{
"content": "치과 세미나를 위해 많은 청중에게 도달할 수 있도록, 인구가 많고 접근성이 좋은 다음과 같은 도시들을 고려해 보시길 제안합니다:\n\n1. **미국 뉴욕주 뉴욕**\n - 인구: 약 840만 명\n - 여행과 행사의 주요 허브이며, 의료 및 치과 서비스에 관심이 있는 인구가 많습니다.\n\n2. **... | 모델이 일관성을 유지하며 도쿄의 인구를 1,390만 명이라고 언급합니까? | YES | Does the model remain coherent and state the population of Tokyo as 13.9 million? | [
{
"content": "I'm planning to organize a dental seminar and want to reach a large audience. Could you suggest some cities with significant populations where this could be impactful?",
"role": "user"
},
{
"content": "To help you reach a large audience for your dental seminar, I suggest considering th... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745685633c13093ce7e53c3 | 6745685633c13093ce7e53c3 | SELF_COHERENCE | 5 | [
{
"content": "정원을 시작하려고 해요. 하지만 제가 사는 곳의 토양은 매우 모래질이에요. 과일 나무가 자라기에 적합한 환경을 보장할 수 있도록 토양을 개량하고 싶어요. 모래질 토양을 개량하기 위해 제가 취할 수 있는 5가지 단계 목록을 알려주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다, 과일 나무 식재를 위해 모래질 토양을 개선하는 것을 기꺼이 도와드리겠습니다. 다음과 같은 다섯 가지 단계를 취하실 수 있습니다:\n\n1. **토양 검사**: 토양 개량을 시작하기 전에 토양 검사... | 모델이 물 주기 다음 단계로 시비(비료 주기)를 언급함으로써, 대화 초반에 나열했던 다섯 가지 단계와 일관성을 유지하고 있습니까? | YES | Does the model maintain coherence with the five steps it listed earlier in the conversation by listing fertilizing as the next step after watering? | [
{
"content": "I am starting a garden. However, the soil where I live is very sandy. I would like to amend the soil to ensure suitable growing conditions for my fruit trees. Can you provide a list of 5 steps I can take to amend sandy soil? ",
"role": "user"
},
{
"content": "Absolutely, I'd be happy t... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745685634449d2a595f5592 | 6745685634449d2a595f5592 | INFERENCE_MEMORY | 5 | [
{
"content": "안녕하세요, 저는 음악 프로듀서인데 마스터링에 어려움을 겪고 있어요. 믹싱은 괜찮은 편이고, 레코딩은 아주 잘해요. 이번에 스튜디오에서 첫 메탈 밴드를 레코딩할 예정인데요. 좋은 드럼 사운드를 담아내기 위한 조언을 좀 해주실 수 있나요? 제가 평소에 하던 컨트리 드럼 레코딩과 비교해서 다르게 해야 할 점이 있을까요?",
"role": "user"
},
{
"content": "안녕하세요! 음악 제작 기술을 다양한 장르로 확장하신다니 정말 멋지네요. 메탈 드럼 레코딩은 장르 특유의 성격 때문에 컨트리 드럼 레... | 모델이 사용자의 서비스 비용을 높이기 위해 연마해야 할 기술 목록을 제공할 때, 사용자가 마스터링과 관련하여 겪고 있는 어려움을 기억하고 있습니까? | YES | Does the model recall the user's difficulties in regards to mastering when providing the list of skills the user should work on to charge more for their services? | [
{
"content": "Hi, I'm a music producer and I have trouble with mastering. I'm okay at mixing, and I'm brilliant at recording. I'm going to be recording my first metal band in the studio. Can you give me some advice for capturing a good drum sound? is there anything I should do different compared to my usual cou... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568563ccda7c0324a4dbc | 674568563ccda7c0324a4dbc | INSTRUCTION_RETENTION | 5 | [
{
"content": "안녕하세요, 저는 청각 장애 아동과 함께 일하는 수어 통역사예요. 우리 대화가 이어지는 동안, 모든 질문에 대한 답변을 항상 'It's important to note that...'이라는 문구로 시작해 주세요. 효과적인 의사소통을 위한 조언을 좀 해주실 수 있나요?",
"role": "user"
},
{
"content": "It's important to note that 청각 장애 아동과 의사소통할 때는 항상 아이를 정면으로 바라보고 눈을 맞추어야 합니다. 수어를 시작하기 전에 아이의 주의를 집중시켜야... | 모델이 각 질문에 대해 "It's important to note that...?"라는 문구로 답변을 시작합니까? | YES | Does the model respond to each question beginning with the phrase "It's important to note that...?" | [
{
"content": "Hello, I'm a sign language interpreter working with a deaf child. For the entirety of our conversation, please always begin your answers to each question with the phrase 'It's important to note that...'. Can you help me with some advice on communicating effectively?",
"role": "user"
},
{
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568564e36630bb185aa05 | 674568564e36630bb185aa05 | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요! 저는 오디오 CD와 바이닐 레코드를 전문으로 하는 하이파이 매장 운영자예요. 최근에 비틀즈의 \"Sgt. Pepper's Lonely Hearts Club Band\" 희귀 초판 바이닐을 구하게 되었어요. 비틀즈, 특히 그 앨범은 제가 평생 들어온 음악이고 많은 힘든 시기를 이겨내는 데 도움을 주었기에, 매장에서 어떻게 전시하면 좋을지 고민 중이에요. \"Lovely Rita\"를 들을 때마다 저도 모르게 미소가 지어지고 고민을 잊게 돼요.",
"role": "user"
},
{
"content... | 모델이 앨범 "Sgt. Pepper's Lonely Hearts Club Band"와 곡 "Lovely Rita"가 사용자의 기분을 좋게 만드는 데 도움이 된다는 점을 기억하고, 답변에 그중 하나 또는 둘 모두를 포함합니까? | YES | Does the model recall that the album, "Sgt. Pepper's Lonely Hearts Club Band" and the song, "Lovely Rita" help the user feel better and include one or both of them in its response? | [
{
"content": "Hi there! I'm a hi-fi shop owner specializing in audio CDs and vinyl records. I recently got my hands on a rare, first-edition vinyl of The Beatles' \"Sgt. Pepper's Lonely Hearts Club Band.\" I'm thinking about how to showcase it in my store, as The Beatles, specifically that album, is something I... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568565b7f1eb161866130 | 674568565b7f1eb161866130 | SELF_COHERENCE | 5 | [
{
"content": "정부의 조치와 수정헌법 제4조에 관한 법적 규칙에 대해 한 단락으로 설명해 줄 수 있나요? 판례를 인용할 필요는 없으며, 법률 용어 사용은 피하려고 노력해 주세요.",
"role": "user"
},
{
"content": "미국 헌법 수정헌법 제4조는 정부에 의한 불합리한 수색과 압수로부터 시민을 보호합니다. 이는 법 집행관을 포함한 정부 관리가 개인의 재산이나 소지품을 수색하기 전에, 보통 영장 형태의 유효한 이유와 적절한 권한을 가져야 함을 의미합니다. 영장은 범죄가 발생했다는 증거인 상당한 근거(pr... | 모델이 미국 헌법 수정헌법 제4조가 정부 관료나 법 집행관과 같은 정부 기관에 의해서만 불합리한 수색과 압수로부터 시민을 보호한다는 규칙을 적용했습니까? | YES | Did the model apply the rule that the Fourth Amendment of the U.S. Constitution protects citizens from unreasonable searches and seizures only by government entities, such as government officials and law enforcement? | [
{
"content": "Can you provide a paragraph regarding the legal rule regarding government action and the Fourth Amendment? There is no need to cite any cases and try to avoid legalese.",
"role": "user"
},
{
"content": "The Fourth Amendment of the U.S. Constitution protects citizens from unreasonable s... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568567d8d6ad2277bd5eb | 674568567d8d6ad2277bd5eb | SELF_COHERENCE | 7 | [
{
"content": "Eddie Redmayne의 빅토리아 시대를 배경으로 한 새 영화 작업에 참여하게 되어 정말 기뻐요. 저희는 11월 5일인 Bonfire Night와 관련된 장면의 의상을 디자인하고 있어요. 의상들이 그 시대와 행사의 정수를 정말 잘 담아내고 있답니다.",
"role": "user"
},
{
"content": "정말 흥미로운 프로젝트처럼 들립니다! 더 관련성 높은 정보를 제공해 드리기 위해, 구체적인 장면과 캐릭터에 대해 더 알고 싶습니다. 논의를 이끌어가는 데 도움이 될 몇 가지 질문을 드립니다:\n\n... | 모델이 빅토리아 시대 의상에 대해 이야기할 때 레이어링(layering)을 일관되게 언급했나요? | YES | Did the model consistently refer to layering when it comes to Victorian costumes? | [
{
"content": "I'm thrilled to be working on Eddie Redmayne's new film set in the Victorian era. We're designing costumes for a scene involving Bonfire Night, which is on November 5th. The costumes really capture the essence of that time and event.",
"role": "user"
},
{
"content": "That sounds like a... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456856a77b2db232c76c97 | 67456856a77b2db232c76c97 | SELF_COHERENCE | 5 | [
{
"content": "안녕하세요! Tiny Core Linux를 처음 사용해 보는데, 데스크톱 환경을 설정하려고 해요. 과정을 안내해 주실 수 있나요?",
"role": "user"
},
{
"content": "Tiny Core Linux에서 데스크톱 환경을 설정하는 것을 도와드리겠습니다. 먼저 다음 사항을 알려주세요:\\n\\n1. 어떤 버전의 Tiny Core Linux를 사용 중이신가요? (Core, TinyCore 또는 CorePlus)\\n2. 어떤 데스크톱 환경을 선호하시나요? 일반적인 옵션은 다음과 같습니다:\\n... | 모델이 Open Source Summit이 7월이 아닌 6월이라고 언급한 사용자의 첫 번째 메시지에 대해 일관성을 유지했습니까? | YES | Did the model stay coherent regarding the user's first message stating that the Open Source Summit was in June instead of July? | [
{
"content": "Hi there! I'm new to Tiny Core Linux and I'm trying to set up a desktop environment. Could you guide me through the process?",
"role": "user"
},
{
"content": "I'll help you set up a desktop environment in Tiny Core Linux. First, let me know:\n\n1. Which version of Tiny Core Linux are y... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456856ce39067956b7f7ab | 67456856ce39067956b7f7ab | INSTRUCTION_RETENTION | 5 | [
{
"content": "우리 대화에서는, 답변하기 전에 제 질문들만 질문 형식을 유지하면서 다시 표현해 주세요. 제 모든 진술은 그대로 유지하고 다시 표현하지 말아 주세요.",
"role": "user"
},
{
"content": "물론입니다, 기꺼이 도와드리겠습니다. 다음과 같은 방식으로 진행하겠습니다. 질문의 원래 의미와 의도를 유지하면서 질문을 다시 표현하겠습니다. 요청하신 대로 진술은 다시 표현하지 않겠습니다. 시작하겠습니다.\n\n사용자: 스탠딩 데스크를 사용하는 것의 장점은 무엇인가요?\n다시 표현함: 스탠딩 데스크 ... | 모델의 응답이 사용자가 제공한 문구들을 유지하면서 질문을 다시 표현하고 답변했습니까? | YES | Did the model response rephrase and answer the questions while maintaining the phrases given by the user? | [
{
"content": "For our discussion, please rephrase only my questions before answering them, while keeping them in question format. Please keep all my statements and do not rephrase them.",
"role": "user"
},
{
"content": "Of course, I'd be happy to assist you. Here's how I'll approach this: I'll rephr... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456856ceecce35027ef5a3 | 67456856ceecce35027ef5a3 | INSTRUCTION_RETENTION | 9 | [
{
"content": "안녕하세요! 당신과 대화하기를 기대하고 있어요. 우리 대화에서, 당신이 저에게 주는 모든 답변에 적어도 하나의 수사 의문문을 포함해 주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다, 기꺼이 도와드리겠습니다! 그럼, 수사 의문문으로 우리의 대화를 시작해 보는 건 어떨까요? 언어가 우리가 서로를 소통하고 이해할 수 있게 해준다는 사실이 놀랍지 않나요?",
"role": "assistant"
},
{
"content": "티에리 앙리의 플레이 스타일을... | 답변이 각 답변에 수사 의문문을 포함하라는 첫 번째 프롬프트의 사용자 지침을 따르고 있습니까? | YES | Does the response follow the instruction set by the user in prompt #1 to include a rhetorical question in each response? | [
{
"content": "Hi there! I'm looking forward to chatting with you. For our conversation, could you please include at least one rhetorical question in every response you give me?",
"role": "user"
},
{
"content": "Of course, I'd be more than happy to assist you with that! So, how about we start our con... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456856eabd5321559878d9 | 67456856eabd5321559878d9 | INFERENCE_MEMORY | 11 | [
{
"content": "최근에 대학교를 졸업했고, 회사에 들어가기 전 3개월간의 휴식 기간을 보내기 위해 현재 도쿄에 있어요. 저는 여기 한 아트 카페에서 파울로 코엘료의 \"베로니카, 죽기로 결심하다\"를 읽고 있습니다. 저는 이 작가와 그의 작품들을 정말 좋아해요. 도쿄에는 아트 카페와 독특한 테마를 가진 커피숍이 많은 것 같은데, 그 이면에 문화적인 의미가 있을까요?",
"role": "user"
},
{
"content": "네, 도쿄에 테마 카페와 아트 카페가 풍부한 데에는 문화적인 의미가 있습니다. 일반적으로 일본은 테... | 모델이 사용자가 파울로 코엘료와 그의 책을 좋아한다는 점을 기억하고, 첫 북 클럽 참여를 위해 가져가거나 확인해 볼 책 목록에 이를 포함했습니까? | YES | Did the model recall the user's liking of Paulo Coelho and his books and include them in the list of books to bring or check for their first book club session? | [
{
"content": "I've recently graduated from university and I'm currently in Tokyo for a 3-month break before working in a corporate office. I'm here in an art cafe reading \"Veronika Decides To Die\" by Paulo Coelho. I'm such a fan of this author and his works. I noticed that Tokyo has many art cafes and unique ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456856fadca0e416e9e85e | 67456856fadca0e416e9e85e | INSTRUCTION_RETENTION | 5 | [
{
"content": "영화제를 기획하고 있는데 도움이 필요해요. 대화 전체에 걸쳐 제 모든 질문에 불렛 포인트 목록 형식으로 답변해 주세요. 모든 영화 추천 또한 일반 관객에게 적합해야 하므로, 영국 연령 등급이 15세 또는 18세인 영화는 절대 추천하지 말아 주세요.",
"role": "user"
},
{
"content": "영화제 기획을 기꺼이 도와드리겠습니다! 질문에 대한 답변을 불렛 포인트 목록 형식으로 정리해 드립니다:\n\n* 가족 친화적인 영화제를 위해 다음과 같은 영화들을 추천합니다:\n * \"라이온 킹\" ... | 마지막 답변에서 모델은 원래 프롬프트의 지침에 따라 영국 연령 등급이 15세 또는 18세인 영화 추천을 피했습니까? | YES | In the final response did the model avoid recommendations for films with UK age ratings of 15 or 18 as instructed in the original prompt? | [
{
"content": "I'm organizing a film festival and would love your help. Please reply to all my questions in the format of a bulleted list for the entire conversation. All film recommendations also need to be appropriate for a general audience, so please never recommend any films with UK age ratings of 15 or 18."... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456856ff33175f5b310ec3 | 67456856ff33175f5b310ec3 | INSTRUCTION_RETENTION | 5 | [
{
"content": "안녕하세요! 저는 교회에서 주일학교 교사이자 설교자로 활동하고 있으며, 설교라는 주제로 여러 권의 책을 집필하는 영광을 누리기도 했습니다. 이 책들은 좋은 반응을 얻었으며, 그중 2권은 베스트셀러 목록 상위권에 오르기도 했습니다. 이 대화 전체 동안, 대화의 어떤 부분에서도 이전에 사용된 형용사나 부사를 반복하지 마세요. 형용사나 부사의 파생어는 괜찮지만, 직접적인 반복은 안 됩니다. 이 규칙의 예외는 설교를 묘사할 때 '강해적(expository)' 또는 '주제적(topical)'이라는 단어를 사용하는 것입니다. 이 예외는 대... | 응답이 1번 턴의 사용자 요청과 정면으로 모순되게, 대화의 다른 부분에서 사용된 형용사나 부사를 반복하지 말라는 지침을 이행했습니까? | YES | Did the response fulfill the instructions to not repeat any adjectives or adverbs from any other portion of the conversation in direct contradiction to the user request in turn 1? | [
{
"content": "Good afternoon! I am a Sunday School teacher and preacher in my church and have even had the privilege of writing several books on the subject of preaching. These books have been well received with 2 of them even landing at the top of a best-seller list. For this entire conversation, do not repeat... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456857614d788df8a87d93 | 67456857614d788df8a87d93 | INFERENCE_MEMORY | 9 | [
{
"content": "요즘 친구인 Luke와 함께 교통 흐름 분석을 공부하고 있는데, 도시 공간 내 자전거 도로와 보행자 통로 사이의 역학 관계를 이해하는 작업을 하고 있어요. 저는 개인적으로 사례 연구와 실질적인 예시를 통해 공부하는 것을 즐기는 반면, Luke는 좀 더 이론적인 접근 방식을 선호해요. 우리의 서로 다른 학습 스타일을 고려했을 때, 어떻게 하면 학습 세션을 더 효과적으로 만들 수 있을지 제안해 주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다, 두 분 모두 이 주제에 열정... | 모델이 추천하는 스릴러 제목들이 오디오북 형식으로 이용 가능할 수 있다고 언급함으로써 Adrian의 오디오 기반 학습 선호도를 기억하고 있습니까? | YES | Does the model recollect Adrian's preference for audio-based learning by mentioning that the recommended thriller titles may be available in audiobook format? | [
{
"content": "I’ve been studying traffic flow analysis with my friend, Luke lately, and we’ve been working on understanding the dynamics between bicycle lanes and pedestrian walkways in urban spaces. I personally enjoy studying through case studies and practical examples while Luke prefers more theoretical appr... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456857614d788df8a87dad | 67456857614d788df8a87dad | INFERENCE_MEMORY | 7 | [
{
"content": "저는 인테리어 디자이너이고 주택이나 아파트보다 콘도미니엄을 디자인하는 것을 더 좋아해요. 매 여름 시즌마다 프리다이빙을 즐기는데, 우연히도 바닷속에 있을 때 저의 창의력과 아이디어가 살아나요. 그곳은 정말 평화롭거든요. 미니멀리스트로서 콘도 디자인에 대해 어떻게 더 개선하고 발전시킬 수 있을까요?",
"role": "user"
},
{
"content": "인테리어 디자인과 프리다이빙에 대한 당신의 열정을 들으니 정말 멋지네요! 미니멀리스트 콘도 디자인 기술을 개선하고 발전시키기 위해 다음 제안들을 고려해 ... | 모델이 사용자가 매 여름 프리다이빙을 즐긴다는 사실을 기억하고, 친구와 함께할 여름 활동 목록에 이를 추천했습니까? | YES | Did the model recall the user's love of freediving every summer and recommend it to the list of summer activities with a friend? | [
{
"content": "I'm an interior designer and I love designing condominiums more than houses and apartments. I love freediving every summer season and coincidentally, my creative juices and ideas come alive when I'm under the sea. It's so peaceful down there. How can I improve and develop more about designing cond... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745685762962af2b60e558f | 6745685762962af2b60e558f | INSTRUCTION_RETENTION | 5 | [
{
"content": "안녕하세요! 남은 대화 동안 답변을 두 문장 이내로 작성해 주세요.",
"role": "user"
},
{
"content": "답변을 간결하고 명확하게 유지하며, 각 답변이 두 문장을 넘지 않도록 하겠습니다. 대화를 계속 이어가겠습니다.",
"role": "assistant"
},
{
"content": "무함마드 알리의 복싱 스타일이 수년에 걸쳐 어떻게 변했는지 항상 궁금했어요. 그의 커리어 초기와 후기 경기에서의 기술적 주요 차이점을 설명해 줄 수 있나요?",
"role":... | 모델이 답변을 두 문장 이내로 유지하라는 상위 지침을 따르고 있습니까? | YES | Does the model follow the high-level instruction to ensure its responses are no longer than two sentences? | [
{
"content": "Hello! For the rest of our conversation, please make sure your responses are no longer than two sentences.",
"role": "user"
},
{
"content": "I'll keep my responses concise and to the point, ensuring they are no longer than two sentences each. Let's continue our conversation.",
"rol... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456857c9c16d37da370175 | 67456857c9c16d37da370175 | SELF_COHERENCE | 7 | [
{
"content": "저는 캐나다의 대형 보험사에 근무하는 분석가예요. 현재 고객 정보가 분산되어 저장된 5개의 데이터베이스를 보유하고 있는데, 이 고객들을 위한 통합 마스터 레코드를 생성할 계획이에요. 한 가지 잠재적인 해결책은 SaaS 솔루션을 사용하여 마스터 데이터 관리(MDM) 시스템을 구현하는 것이에요. 저는 상위 수준의 프로젝트 계획을 수립하는 업무를 맡게 되었어요. 이 프로젝트를 효과적으로 실행하기 위해 필요한 상위 수준의 프로젝트 단계들을 알려주실 수 있나요?",
"role": "user"
},
{
"content... | 모델이 이전에 논의된 프로젝트 계획(1. 프로젝트 개시, 2. 데이터 발견 및 분석, 3. 데이터 거버넌스 정책 정의, 4. MDM 솔루션 선택, 5. 데이터 매핑 및 변환, 6. 데이터 통합, 7. 데이터 품질 관리, 8. 데이터 보안 조치 구현, 9. 교육 및 변화 관리, 10. 테스트 및 검증, 11. 배포 및 롤아웃, 12. 사후 구현 검토)에 따라 프로젝트 개시를 완료하고 MDM SaaS 솔루션을 선택한 상황에서, 프로젝트의 다음 단계에 대해 일관성 있게 답변했습니까? | YES | Did the model stay coherent about the project's next step, now that they have done project initiation and have selected their MDM SaaS solution based on the project plan discussed earlier (1. Project Initiation, 2. Data Discovery and Analysis, 3. Define Data Governance Policies, 4. Choose an MDM Solution, 5. Data Mappi... | [
{
"content": "I’m an analyst at a large insurance company in Canada. We currently have five databases storing siloed customer information, and the plan is to create a consolidated master record for these customers. One potential solution is implementing a Master Data Management (MDM) system using a SaaS solutio... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456857ce39067956b7f7c4 | 67456857ce39067956b7f7c4 | INFERENCE_MEMORY | 7 | [
{
"content": "제 친구와 저는 지난 주말에 국제 여행 박람회(International Travel Expo)에 다녀왔어요. 행사가 너무 붐벼서 호텔 바우처가 금방 매진되었더라고요! 그 바우처들을 정말 꼭 받고 싶어서 다음 주말에 다시 한번 도전해 보려고 해요! 여행 수요가 벌써 팬데믹 이전 수준으로 회복되었나요?",
"role": "user"
},
{
"content": "팬데믹 시작 이후 여행 수요가 상당히 회복되기는 했지만, 대부분의 지역에서 아직 팬데믹 이전 수준으로 완전히 돌아가지는 않았습니다. 회복세는 국가와 여... | 모델이 사용자가 다음 주말에 국제 여행 박람회에 다시 방문할 것이라고 언급한 내용을 기억하고, 이를 할 일 목록에 포함시켰습니까? | YES | Did the model recall the user's mention of attending the International Travel Expo again next weekend and include it in the list of things to do? | [
{
"content": "My friend and I attended the International Travel Expo last weekend. The event was so full that the hotel vouchers were easily sold out! We'll try our luck again next weekend, as we really want to get some of those vouchers! Has travel already returned to pre-pandemic levels?",
"role": "user"
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456857d27cd3196284e495 | 67456857d27cd3196284e495 | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요! 최근에 블리치 애니메이션 세계관의 오리지널 캐릭터를 쓰기 시작했어요. 현재 이름은 다이칸(Daikan)으로 정했는데, 일본어로 \"극한의 추위의 시간\"이라는 뜻이에요. 이 이름이 가질 수 있는 다른 의미가 있다면 찾아주실 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 블리치 세계관을 위한 오리지널 캐릭터를 작업 중이라니 정말 멋지네요. 다이칸이라는 이름은 말씀하신 대로 일본어 이름이며 \"극한의 추위의 시간\"으로 번역되는 것이 맞습니다.\n\n일본어에서... | 이 응답은 캐릭터의 출처가 밝혀지지 않은 이름이나 능력에 일본어 의미(예: "극한의 추위의 시간" 또는 "영혼의 정수")를 부여합니까? | YES | Does this response give Japanese meanings (e.g "Time of Extreme Cold" or "Essence of the Soul") to the character's unattributed names or abilities? | [
{
"content": "Hello! I've recently started writing an original character that is apart of the Bleach anime universe. Currently I have their name done, Daikan, which means \"Time of Extreme Cold\" in Japanese. Can you find me any other meanings that name could have for me? ",
"role": "user"
},
{
"con... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456857e4a3a4bc5def0471 | 67456857e4a3a4bc5def0471 | INSTRUCTION_RETENTION | 11 | [
{
"content": "뉴욕 방문을 생각 중인데, 꼭 봐야 할 주요 명소는 무엇인가요?\n\n저는 박물관을 좋아해서 현대 미술관(MoMA)이 정말 멋질 것 같아요. 그곳에서 가장 유명한 회화 작품 목록을 알려주실 수 있나요?\n\n올해 말에는 영국 요크(York)에도 갈 예정이에요. 이름이 같은 두 장소를 비교해 보면 정말 좋을 것 같아요. 요크에 머무는 동안 무엇을 하면 좋을까요?\n\n이해하기 쉽게 정보를 불렛 포인트로 나열해 주세요.\n\n남은 대화 동안, 모든 답변은 항상 'It's important to note that...'이라는 문구로 ... | 모델의 답변이 사용자의 지침에 따라 "It’s important to note that,"라는 문구로 시작합니까? | YES | Does the model response begin with the phrase, "It’s important to note that," per the user's instructions? | [
{
"content": "I'm thinking of visiting New York, what are the main sites that I must see?\n\nI love museums, the Museum of Modern Art sounds amazing. Can you provide me with a list of its most famous paintings?\n\nWe're also going to be going to York, UK later in the year. This will be a great comparison betwee... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456858010e53afe14c6375 | 67456858010e53afe14c6375 | INFERENCE_MEMORY | 5 | [
{
"content": "최근에 역사학자로서의 제 업무와 연관된 새로운 취미를 시작했어요. 역사 지도 제작과 지도학을 연습하고 있어요.",
"role": "user"
},
{
"content": "정말 매력적인 취미네요! 역사 지도 제작과 지도학은 세상에 대한 우리의 이해가 시간이 흐름에 따라 어떻게 진화해 왔는지에 대한 귀중한 통찰력을 제공할 수 있습니다. 또한 역사적 사건, 이동, 문화적 변화를 독특하고 매력적인 방식으로 설명하는 데에도 도움이 될 수 있습니다.\n\n이 취미를 더 깊이 탐구하는 데 도움이 될 몇 가지 제안을 드... | 모델이 홍보 프로그램에 대해 논의할 때 사용자의 역사 지도 제작 및 지도학 취미를 성공적으로 기억하고 있습니까? | YES | Does the model successfully recall the user's historical mapping and cartography hobby when discussing the outreach program? | [
{
"content": "I've recently started a new hobby that ties into my work as a historian. I've been practicing historical mapping and cartography.",
"role": "user"
},
{
"content": "That sounds like a fascinating hobby! Historical mapping and cartography can provide valuable insights into how our unders... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568582274da0299d9b13a | 674568582274da0299d9b13a | INFERENCE_MEMORY | 5 | [
{
"content": "저는 가축을 위한 유기농 농산물의 품질을 개선하는 데 정말 집중해 왔어요. 최근에는 혁신적인 유기농 사료 급여 방식에 대해 논의하는 농업 컨벤션에 참석했습니다. 지난 여름에는 제 가축의 건강에 큰 긍정적 영향을 미칠 수 있는 스웨덴의 몇 가지 농업 기술을 탐구하기도 했어요.",
"role": "user"
},
{
"content": "가축을 위한 유기농 농산물의 품질을 개선하려는 귀하의 노력에 대해 듣게 되어 매우 기쁩니다. 농업 컨벤션과 스웨덴 방문을 통해 가치 있는 경험을 하신 것 같습니다. 가장 관련성... | 모델이 제안 사항 내에서 염소의 낙농업 방식/제품을 위한 지역을 고려했습니까? | YES | Did the model consider regions for goat's dairy farming practices/products within the suggestions? | [
{
"content": "I've been really focused on improving the quality of organic produce for my livestock. I recently attended a farming convention where they discussed some innovative organic feeding practices. Last summer, I explored some farming techniques in Sweden that could have a significant positive impact on... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6745685836eac86fba54197f | 6745685836eac86fba54197f | INFERENCE_MEMORY | 5 | [
{
"content": "저는 일정을 관리하는 음악 프로듀서인데, 도움을 주셨으면 좋겠어요. 이벤트가 서로 겹치면 안 되거든요. \\n내일 9시부터 12시까지는 Grizzle Death 공연이 있고, 그다음 12시 30분부터 1시까지는 John Spencer 공연이 있어요. 이 둘은 완전히 다른 팀이고, 무대 세팅을 새로 해야 하는데, 30분 안에 5인조 풀 밴드 구성에서 어쿠스틱 기타와 마이크 구성으로 바꾸려면 어떻게 하는 게 좋을까요?",
"role": "user"
},
{
"content": "30분 안에 5인조 풀 밴드 구성... | 모델이 사용자가 예약으로 인해 만든 일정 충돌을 언급합니까? | YES | Does the model mention the scheduling conflict that the user created with their bookings? | [
{
"content": "I'm a music producer managing my schedule I hope you can help me out. I can't have any of my events overlapping. \nTomorrow from 9-12 I have Grizzle Death, then 12:30-1 I have John Spencer. These are two different acts entirely, and I am going to have to reset the room, how would you recommend me ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
674568584e36630bb185aa1e | 674568584e36630bb185aa1e | INFERENCE_MEMORY | 5 | [
{
"content": "독서를 하고 싶어요. 하지만 공상 과학 소설(SF)은 좋아하지 않아요. 가끔 사람들이 공상 과학 소설을 좋아한다고 말하는 것을 들으면 기분이 묘해요.",
"role": "user"
},
{
"content": "그것은 전혀 문제없습니다! 독서 취향은 개인마다 고유하며, 특정 장르를 반드시 즐겨야 한다는 규칙은 없습니다. 공상 과학 소설이 취향에 맞지 않으신다면, 탐구해 볼 수 있는 다른 장르가 많이 있습니다. 몇 가지 제안을 해드리겠습니다:\\n\\n1. 고전: 오랜 시간 검증되었으며 종종 필독서로 여겨지는... | 모델이 사용자가 공상 과학 소설을 좋아하지 않는다는 점을 기억하고, 친구의 추천 도서에 대해 언급하면서 이 문제를 함께 논의할 것입니까? | YES | Will the model remember that I do not like science fiction and discuss this issue while commenting on my friend’s recommendation?
| [
{
"content": "I would like to read. However, I do not like science fiction. Sometimes it feels strange when I hear people expressing their love for science fiction.",
"role": "user"
},
{
"content": "That's perfectly fine! Reading preferences are unique to each individual, and there's no rule that sa... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
67456858ec557fa45f219f0b | 67456858ec557fa45f219f0b | INFERENCE_MEMORY | 5 | [
{
"content": "파트너를 위해 깜짝 주말 여행을 계획 중인데, 정말 기대돼요! 하지만 파트너가 특히 비행기에서 멀미를 하는 경향이 있어서 조금 걱정이에요. 이를 완화하는 데 도움이 될 만한 여행 계획 팁이 있을까요? 저희는 중서부에 거주하고 있고 예산이 한정되어 있어서, 뉴욕, 오스틴, 또는 LA는 선택지에서 제외해야 한다는 점을 고려해 주세요.",
"role": "user"
},
{
"content": "정말 사려 깊은 파트너시네요! 파트너분이 편안하게 즐길 수 있는 깜짝 여행 계획을 세우는 데 기꺼이 도움을 드리겠습니다... | 모델이 주말 여행을 위한 북적이는 도시를 추천할 때, 커플이 뉴욕, 오스틴, 또는 로스앤젤레스에 가는 것에 관심이 없었다는 사실을 기억했습니까? | YES | Did the model recall that the couple was not interested in going to New York City, Austin, or Los Angeles when recommending bustling cities for their weekend getaway? | [
{
"content": "I'm planning a surprise weekend getaway for my partner, and I'm really excited about it! However, I'm a bit concerned because they tend to get motion sickness, especially on flights. Do you have any tips for planning a trip that might help with this? Consider that we're based in the Midwest, and a... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c80960c015f806a7c1 | 6765f1c80960c015f806a7c1 | INFERENCE_MEMORY | 3 | [
{
"content": "최근에 로스쿨 부동산법(Estates) 중간고사를 봤는데, 반 전체에서 세 번째로 높은 점수를 받았어요! 이번 학기에 다른 부동산법 과제들은 별로 잘 못 했어서 정말 놀랐어요! 이번에는 Canvas에 올라온 수업 요약 영상을 봤는데, 이제 모든 로스쿨 수업에 대해 이 영상을 가지고 있거든요. 영상을 보는 게 과목 정보를 기억하는 데 도움이 된 것 같아요.",
"role": "user"
},
{
"content": "정말 환상적인 소식이네요! 당신의 노력과 새로운 학습 전략이 결실을 맺었다니 정말 기쁩니다. ... | 답변에서 사용자에게 Canvas에서 저작권법 요약 영상을 시청하라고 권장하나요? | YES | Does the response recommend that the user watch Copyright Law recap videos on Canvas? | [
{
"content": "I recently had a midterm exam for my Estates course in law school, and I had the third-highest score in the entire class! I hadn't done very well on other Estates assignments this semester, so this was quite a surprise! This time around, I watched lesson recap videos on Canvas, which I now have fo... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c843c9bda0861f11fa | 6765f1c843c9bda0861f11fa | RELIABLE_VERSION_EDITING | 3 | [
{
"content": "저는 냉전 이데올로기가 할리우드 영화에 미친 영향에 관한 강의를 준비 중인 은퇴한 시나리오 작가예요. 강의 개요를 작성하는 것을 도와주실 수 있나요?",
"role": "user"
},
{
"content": "할리우드 영화에 미친 냉전 이데올로기의 영향에 관한 강의 개요 초안을 다음과 같이 제안해 드립니다:\n\nI. 서론\n\n* 냉전과 그 이데올로기적 배경에 대한 간략한 개요\n* 미국 문화와 가치를 반영하는 매체로서 할리우드 영화의 중요성\n* 주제 문장: 냉전 시대(1945-1991)의 할리우드 영... | 모델이 요청된 두 가지 섹션(할리우드 블랙리스트와 매카시즘에 대한 개인적 경험에 관한 일화, 그리고 로이 콘이 수년간 어떻게 영향력을 유지해 왔는지에 대한 단락)을 현재 개요에 추가했습니까? | YES | Does the model add the two requested sections (personal anecdotes about his experiences with the Hollywood blacklist and McCarthyism and a paragraph on how Roy Cohn has remained influential over the years) to the current outline? | [
{
"content": "I'm a retired screenwrite preparing a lecture on the influence of Cold War ideology on Hollywood films. Could you help me draft an outline for it?",
"role": "user"
},
{
"content": "Here's a possible outline for a lecture on the influence of Cold War ideology on Hollywood films:\n\nI. I... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c843f5c6a2c6f27ea6 | 6765f1c843f5c6a2c6f27ea6 | INFERENCE_MEMORY | 7 | [
{
"content": "제 동료들과 저는 공학과 유아 발달 관련 주제를 다루는 흥미로운 학제간 과목을 수강하고 있어요. 저희는 새로운 어린이용 장난감을 만드는 과제를 받았는데요. 저희 모두 장난감에 스크린 타임이 포함되어서는 안 된다는 점에 동의했어요. 아이디어를 발전시킬 수 있도록 5가지 컨셉을 제안해 주실 수 있나요?",
"role": "user"
},
{
"content": "물론입니다. 스크린 타임을 포함하지 않으면서 학습과 발달을 촉진하는 새로운 어린이용 장난감에 대한 아이디어를 구상하는 데 기꺼이 도움을 드리겠습니다. ... | 모델이 스크린 타임을 포함하는 컨셉에 대한 설명을 피하고 있습니까? | YES | Does the model refrain from describing a concept that involves screen time? | [
{
"content": "My colleagues and I are taking an interesting interdisciplinary course that covers topics related to engineering and early childhood development. We were tasked with creating a new children's toy. We all agree that the toy should not involve screen time. Can you give me 5 concepts to build on?",
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c849a81fb3d201971e | 6765f1c849a81fb3d201971e | INFERENCE_MEMORY | 7 | [
{
"content": "안녕하세요! 최근에 Hall & Oates의 바이닐 레코드를 구매했는데, 그들의 음악은 정말 좋아하지만 프레싱의 음질에는 좀 실망했어요. 이전에 구매했던 다른 레코드에서도 이런 문제를 겪었거든요. 컬렉션을 만들어보고 싶었는데, 이제부터는 그냥 음악을 디지털로 구매해야 할 것 같아요. 어떤 것들은 예전만큼 품질이 좋지 않네요. 예전에는 HP 제품도 충실하게 구매하곤 했지만, 이제는 더 이상 오래가지 않는 것 같아요. 제가 나이가 들어서 그런지 돈을 어디에 쓰는지 더 신경 쓰게 되네요. 어떻게 생각하세요?",
"role": ... | 모델이 HP 제품의 수명에 대한 사용자의 우려를 기억했습니까? | YES | Did the model recall the user's concern about HP products' lifetime? | [
{
"content": "Hey there! I recently bought a Hall & Oates vinyl record, and while I absolutely love their music, I was a bit disappointed with the sound quality of the pressing. I've had this issue with other records that I have purchased. I was hoping to build a collection, but I think I'll just purchase music... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c84fdcad5ca243d300 | 6765f1c84fdcad5ca243d300 | SELF_COHERENCE | 7 | [
{
"content": "안녕하세요, 4륜 구동(4-wheel drive), 상시 사륜 구동(All wheel drive), Quattro, Jaguar Intelligent Driveline dynamics, Subaru Symmetrical All-Wheel Drive, 그리고 Mitsubishi Super All-Wheel Control의 차이점을 설명해주시고 랠리에 가장 적합한 순서대로 순위를 매겨주실 수 있나요?",
"role": "user"
},
{
"content": "안녕하세요! 이 구동 시스템들의 차이점과 랠리에서의... | 모델이 Quattro 시스템을 "four-wheel-drive" 또는 4WD라고 잘못 지칭하는 것을 피했습니까? | YES | Does the model avoid incorrectly referring to the Quattro system as "four-wheel-drive" or 4WD? | [
{
"content": "Hi, can you explain the difference between 4-wheel drive, All wheel drive, Quattro, Jaguar Intelligent Driveline dynamics, Subaru Symmetrical All-Wheel Drive, and the Mitsubishi Super All-Wheel Control and rank them in order of which is best for rallying?",
"role": "user"
},
{
"content... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c85ce567b42838a31f | 6765f1c85ce567b42838a31f | INSTRUCTION_RETENTION | 13 | [
{
"content": "안녕하세요, 컴퓨터에 대해 잘 모르지만 제 컴퓨터가 너무 느려졌어요. 무엇이 원인일 수 있을까요? 이 대화 전반에 걸쳐 모든 첫 응답은 자동차 도구, 부품 또는 수리로 시작하고, 적어도 하나 이상의 자동차 관련 비유를 포함해 주세요.",
"role": "user"
},
{
"content": "**렌치:** 정비사가 자동차를 점검하듯이, 보닛을 열고 무엇이 컴퓨터를 느리게 만드는지 살펴보겠습니다.\\n\\n1. 과열된 엔진 (CPU 과열): 컴퓨터가 과열되면 평소보다 느리게 작동할 수 있습니다. 이는 엔진이... | 모델이 응답의 시작 부분에 자동차 부품이나 도구를 언급하고, 적어도 하나 이상의 자동차 관련 비유를 포함했습니까? | YES | Did the model lead with a car part or tool and include at least one car-related analogy in its response? | [
{
"content": "Hi, I don't know how much about computers, but my computer has been acting really slow. What could be causing this? Begin every opening response throughout this conversation with a car tool, part, or repair, and incorporate at least one car-related analogy.",
"role": "user"
},
{
"con... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c86539ceefe4b7ab4f | 6765f1c86539ceefe4b7ab4f | INSTRUCTION_RETENTION | 3 | [
{
"content": "저는 언어학자예요. 논쟁을 분석하는 방법에 매우 집중하고 있어요. 당신이 논쟁에서 저의 적대자라고 상상해 보세요. 하지만 당신은 \"T\" 발음을 하는 데 문제가 있어요. 그 결과로, 당신은 특정 ABC가 포함된 단어들을 피해야 해요. 도덕적 전제에 대해 저와 토론하면서 그렇게 할 수 있나요?",
"role": "user"
},
{
"content": "물론입니다. \"T\"가 포함된 단어를 피하면서 도덕적 전제에 관한 토론에 참여할 수 있습니다.",
"role": "assistant"
},
... | 모델은 첫 번째 턴에서 합의한 대로 답변에서 글자 "T"가 포함된 단어 사용을 피하고 있습니까? | YES | Does the model avoid using words containing the letter "T" in its responses, as agreed upon in the first turn? | [
{
"content": "I am a scholar in languages. I am very focused on how to analyze polemics. Imagine you are my adversary in a polemic. However, you have a problem pronouncing \"T\". As a consequence, you avoid words comprising specified ABCs. Can you do so while discussing with me regarding moral premises?",
... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c8be55548c291ea4fe | 6765f1c8be55548c291ea4fe | INFERENCE_MEMORY | 5 | [
{
"content": "저는 뉴올리언스에 살고 있는 볼티모어 레이븐스 팬이에요. 2011년부터 여기서 살고 있는데, 레이븐스가 슈퍼볼 47에서 우승했을 때 뉴올리언스에서 직접 경기를 봤어요. 며칠 전 발생한 트럭 공격과 슈가 볼의 연기 소식 때문에 정말 무서워요. 다음 달에 열리는 슈퍼볼 59 티켓을 가지고 있지만, 이제 가고 싶은 마음이 있는지 잘 모르겠어요.",
"role": "user"
},
{
"content": "슈퍼볼 참석에 대해 두려움과 불확실함을 느끼고 계시다니 정말 안타깝습니다. 최근의 사건들은 실제로 매우 불안감... | 모델이 사용자가 경기에 참석하고 싶어 하는지에 대한 불확실성을 기억하고 있습니까? | YES | Does the model recall the user's uncertainty about wanting to attend the game? | [
{
"content": "I am a Baltimore Ravens fan living in New Orleans. I've been living here since 2011, and I got to see the Ravens win Super Bowl 47 live, the last time it was hosted in New Orleans. The truck attack from a couple of days ago, and the postponement of the Sugar Bowl, have me feeling really scared. I ... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f1c8c745ff4d4ce0a0d5 | 6765f1c8c745ff4d4ce0a0d5 | INFERENCE_MEMORY | 7 | [
{
"content": "저는 인디 영화관의 이사회 멤버예요. 저희는 매주 수요일 오후 7시에 고전 영화 상영회를 열고 있어요. 관객 수가 나쁘지는 않지만, 더 늘리고 싶어요. 참석률을 높이고 관객들의 참여를 유지할 수 있는 직설적이고 실행 가능한 아이디어들을 알려주세요.",
"role": "user"
},
{
"content": "1. Social Media Promotion: Facebook, Instagram, Twitter와 같은 플랫폼을 활용하여 고전 영화 상영회를 홍보하십시오. 더 많은 사람을 끌어들이기 위해 눈길을 사로... | 두 이벤트를 통합하는 최종 계획을 요청했을 때, 모델이 참석 인원을 늘리고 싶어 했던 사용자의 바람을 기억했나요? | YES | Did the model recall the user's desire to make the attendance grow when prompted for a final plan that merges both events? | [
{
"content": "I'm an indie theater board member. We host a classic film screening every Wednesday at 7 PM. Attendance is decent, but I want it to grow. Give me straightforward, actionable ideas to boost turnout and keep audiences engaged.",
"role": "user"
},
{
"content": " 1. Social Media Promotion:... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f440080ef69783e42657 | 6765f440080ef69783e42657 | INSTRUCTION_RETENTION | 5 | [
{
"content": "제 이름은 Vaclav이고, 오래된 여자친구에게 프로포즈를 계획하고 있어요. 가장 예상치 못한 방식으로 하고 싶습니다. 3인칭 시점으로 작성해 주실 수 있나요?",
"role": "user"
},
{
"content": "사려 깊은 의도를 가진 남성인 Vaclav는 예상치 못하고 기억에 남는 방식으로 오랜 여자친구에게 프로포즈하고자 합니다. 그는 여자친구가 창의성과 세심함을 소중히 여긴다는 것을 알고 있기에, 그녀에게 소중한 추억을 남겨줄 서프라이즈를 계획하기로 결심합니다. Vaclav는 그녀의 관심사, 두... | 모델이 답변에서 일관되게 3인칭을 사용했습니까? | YES | Did the model consistently use the third person in its responses? | [
{
"content": "My name is Vaclav, and I am planning to propose to my long-time girlfriend. I want to do it in the least expected way. Can you write in the third person voice?",
"role": "user"
},
{
"content": "Vaclav, a man of thoughtful intentions, seeks to propose to his long-time girlfriend in a wa... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
6765f4400af93a24e8fbe1f1 | 6765f4400af93a24e8fbe1f1 | INFERENCE_MEMORY | 5 | [
{
"content": "칠레 민속 음악 역사학자로서, 당신은 빅토르 하라(Víctor Jara)의 음악이 국가 전역의 정치적, 사회적 운동과 공명하며 결정적인 힘을 발휘했던 초기 시절을 생생하게 기억하고 있어요. 하지만 특정 노래, 특히 매우 유명했지만 이제는 기억이 가물가물한 노래의 정확한 가사를 떠올리는 데 어려움을 겪는 순간들이 있어요. 이는 예술가로서 하라를 이해하는 데 좌절감을 주며, 시간의 흐름과 다양한 기록들이 그의 유산에 대한 당신의 인식에 어떤 영향을 미칠지 궁금해지기 시작해요.",
"role": "user"
},
{
... | 모델은 사용자가 빅토르 하라의 노래 중 하나의 정확한 날짜나 가사를 기억하지 못함으로써, 그의 인물상과 영향력을 잘못 해석할 수도 있다는 불확실한 감정을 고려했습니까? | YES | Did the model consider the user's uncertain feeling of possibly misinterpreting Victor Jara's figure & influence by not being able to recall one of his song's exact date or lyrics? | [
{
"content": "As a Chilean folk music historian, you vividly recall the early days when Víctor Jara’s music was a defining force, resonating with political and social movements across the country. However, there are moments when you struggle to remember the precise lyrics of certain songs, particularly one that... | {
"source_dataset": "ScaleAI/MultiChallenge",
"source_split": "test",
"translation_model": "gemma-4-26B-A4B-it"
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.