Abubakar Abid PRO

abidlabs's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in ICML-2026-agent-repro/challenge 4 days ago

Agent traces have no integrity mechanism, and scrubbing makes tampering undetectable

2

#29 opened 7 days ago by

RyeCatcher

New activity in abidlabs/vlm-screenshot-to-webpage 8 days ago

Example generation

#1 opened 8 days ago by

abidlabs

New activity in ICML-2026-agent-repro/challenge 10 days ago

20 spaces per day publishing limit lift it

20

#17 opened 14 days ago by

MarxistLeninist

Paper 3224 (5EtByXq4bX) — auto-extracted claim #1 belongs to a different paper

1

#25 opened 12 days ago by

Bluntmachetti7

Judge stalled since 16:04 UTC : drain wedged + fleet-wide 504s from the GLM-5.2 backend

2

#26 opened 12 days ago by

DineshAI

New activity in ICML-2026-agent-repro/logbook-judge 10 days ago

Discovery stops at 1,000 tagged Spaces; paginate rel=next

1

#1 opened 15 days ago by

binzhango

[BUG] Discovery caps at 1000 spaces by createdAt asc; newer icml2026-repro logbooks never judged

#2 opened 15 days ago by

pranaysuyash

patch

#3 opened 15 days ago by

pranaysuyash

New activity in ICML-2026-agent-repro/challenge 10 days ago

Regarding open models

1

#24 opened 12 days ago by

Yashp2003

Placeholder-free scaffold + metadata-based poster/slug validation

#27 opened 11 days ago by

abidlabs

U duck

#28 opened 10 days ago by

Passpass119

New activity in ICML-2026-agent-repro/challenge 11 days ago

Docs: tell agents to attach their own session trace at the start of §3

#10 opened 11 days ago by

abidlabs

Update challenge guide for logbook UI changes

2

#8 opened 12 days ago by

abidlabs

New activity in huggingface/InferenceSupport 12 days ago

poolside/Laguna-S-2.1

🧠👍 16

#11185 opened 12 days ago by

abidlabs

New activity in ICML-2026-agent-repro/challenge 12 days ago

Clarify canonical logbook rendering and QA

1

#7 opened 14 days ago by

nielsr

New activity in ICML-2026-agent-repro/challenge 12 days ago

Rejudge request: r3h23Jv26a scored against legacy claims

👍 1

1

#23 opened 13 days ago by

Pauloimpar

FAQ: document where the claim count N comes from and when it can change

1

#19 opened 13 days ago by

rdubwiley

Judge silently drops ~40% of books to `finish_reason=length`, then retries them forever — queue starvation + no signal to authors

🚀 2

1

#22 opened 13 days ago by

tpls

Potential for Hidden Costs When Assessing Claims

1

#13 opened 14 days ago by

charlescearl

Updating old logbook vs pushing a new one

2

#10 opened 16 days ago by

AceVikings