Multilingual GSM-Symbolic: What determines capability transfer across languages?
Paper • 2610.03367 • Published • 47
Sentinel
Sentinel7
P(doom) Rejects the premise
AI & ML interests
None yet
Recent Activity
upvoted a paper about 5 hours ago
Multilingual GSM-Symbolic: What determines capability transfer across languages? upvoted a paper about 5 hours ago
Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory upvoted a paper about 5 hours ago
EditHero: A Benchmark for Long-Horizon Part-Level 3D Editing and Vibe ModelingOrganizations
None yet
upvoted 10
papers about 5 hours ago
Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory
Paper • 2610.02521 • Published • 49
EditHero: A Benchmark for Long-Horizon Part-Level 3D Editing and Vibe Modeling
Paper • 2610.02298 • Published • 50
ALoDLM: Adaptively Looped Diffusion Language Models
Paper • 2610.04198 • Published • 60
Latent-MOPD: Latent Multi-Teacher On-Policy Distillation
Paper • 2610.02381 • Published • 63
On-Policy Parameter Update Direction Underlies Generalization in LLM Post-Training
Paper • 2609.36659 • Published • 81
Native Action-Prior Learning from Videos for World Action Models
Paper • 2610.03391 • Published • 81
World Action Modeling with Progressive Visual Planning
Paper • 2610.02508 • Published • 83
Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite
Paper • 2610.02826 • Published • 91
RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations
Paper • 2610.01780 • Published • 268
upvoted 10
papers 1 day ago
PixelUMM: Encoder-Free Unified Image and Video Understanding and Generation
Paper • 2609.38597 • Published • 32
Ego2Act: Evaluating Goal-Directed Manipulation in Egocentric Video Generation
Paper • 2610.01092 • Published • 33
DC-SAE: Deep Compression Semantic Autoencoder for Faster Diffusion Convergence
Paper • 2609.39222 • Published • 42
It's Not What the Image Shows: Irrelevant Context Destabilises VLM Judges Without Informing Them
Paper • 2609.37863 • Published • 38
AutoDataBench: A Data-centric Testbed for Accelerating Auto Research
Paper • 2609.40097 • Published • 45
Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models
Paper • 2609.33355 • Published • 50
4Director: Controlling Video World Models with Rigid 3D Geometry
Paper • 2610.02160 • Published • 38
SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation
Paper • 2610.02201 • Published • 36
FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation
Paper • 2609.38839 • Published • 93
Agent Error Dataset: Scaling 50,000 Error--Diagnosis Pairs for Failure Analysis and Error-Aware Post-Training
Paper • 2609.40111 • Published • 50