Good Papers

Showing papers from Meta / Ecole Normale Supérieure Show all papers

88%Must read
?Must readVote to see the score
NeurIPS 2026Meta / Ecole Normale SupérieureU California, San DiegoINRIA - Ecole Normale SuperieureFacebook AI ResearchMeta / FAIRRL for LLMs

Don’t Let Gains FADE: Breaking Down Policy Gradient Weights in RL

A framework decomposes RL advantage functions into gradient mass axes, showing trade-offs shift during training and motivating FADE, which adapts weights dynamically to accelerate convergence and improve accuracy-diversity trade-offs.

Juliette Decugis, Sean O'Brien, Francis Bach, Gabriel Synnaeve and 1 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

– ReadersNo votes yet
15/20 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: 15 of 20 reviewers recommend it
lenient 4/5
medium 10/10
strict 1/5
92%Must read
?Must readVote to see the score
NeurIPS 2026Meta AIMeta FAIR & Université ParisMeta / Ecole Normale SupérieureInriaFacebook AI ResearchRL for LLMs

Reinforcement Learning for Code Optimization

Reinforcement learning for code optimization fails due to noisy, sparse execution-time rewards, so a calibrated three-stage pipeline improves strict pass rates by up to 125% while preserving correctness.

Pierre Chambon, Kunhao Zheng, Juliette Decugis, Benoît Sagot and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 13 on Hugging Face

– ReadersNo votes yet
19/20 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: 19 of 20 reviewers recommend it
lenient 4/5
medium 10/10
strict 5/5
91%Must read
?Must readVote to see the score
NeurIPS 2026Meta FAIR & Université ParisMeta / Ecole Normale SupérieureMeta AIMeta AIMeta / FAIRRL for LLMs

Extrapolative Weight Averaging Reveals Correctness–Efficiency Frontiers in Code RL

Nested unit-test coverage in code RL reveals a correctness, efficiency frontier that extrapolative weight averaging extends, enabling complementary checkpoints that improve pass@250 by 3.3%.

Kunhao Zheng, Juliette Decugis, Pierre Chambon, Jonas Gehring and 3 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

– 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 5/5
medium 8/10
strict 4/5