Good Papers

Showing papers from New York University; Anthropic Show all papers

57%Worth a look
?Worth a lookVote to see the score
NeurIPS 2026New YorkUCLAU Illinois at Urbana-ChampaignColumbiaNYULLM reasoning & chain of thought

Understanding Reasoning from Pretraining to Post-Training: Chess as a Controlled Testbed

Jingyan Shen, Ang Li, Salman Rahman, Yifan Sun and 3 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026

– ReadersNo votes yet
1/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 1 of 20 reviewers recommend it
lenient 1/5
medium 0/10
strict 0/5
91%Must read
?Must readVote to see the score
NeurIPS 2026PrincetonAnthropicNew YorkNew York University; AnthropicMITLLM evaluation & benchmarks

DiscoverPhysics: Benchmarking LLMs for out-of-the-box scientific thinking

DiscoverPhysics benchmarks LLM agents on simulated worlds with non-standard physics, finding frontier models pass only half and fail at uncovering latent structure.

Lindsay Smith, Matt Sampson, Siddharth Mishra-Sharma, Peter Melchior and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 1 on Hugging Face

– ReadersNo votes yet
17/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 17 of 20 reviewers recommend it
lenient 4/5
medium 9/10
strict 4/5
78%Highly rated
?Highly ratedVote to see the score
NeurIPS 2026École PolytechniqueAMIAD and Safran TechSafranValeoNew York University; AnthropicLLM evaluation & benchmarks

Position: Semantic Uncertainty Measures Disagreement, Not Reliability

Semantic uncertainty measures answer disagreement rather than reliability, as valid answers vary and repeated errors appear certain; a bias-uncertainty decomposition separates variability from systematic error to improve evaluation.

Joseph Hoche, Maxime Corlay, David Brellmann, Andrei Bursuc and 3 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
11/20 AI panelreviewers recommend it

Readers and the AI panel: vote on this paper to see what they said.

Only vote on papers you've read. Sign in with GitHub to vote.

AI panel: 11 of 20 reviewers recommend it
lenient 5/5
medium 5/10
strict 1/5