Cover Image

Extrapolative Weight Averaging Reveals Correctness-Efficiency Frontiers in Code RL

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zheng, Kunhao, Chambon, Pierre, Decugis, Juliette, Gehring, Jonas, Cohen, Taco, Negrevergne, Benjamin, Synnaeve, Gabriel
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

What Makes Large Language Models Reason in (Multi-Turn) Code Generation?
by: Zheng, Kunhao, et al.
Published: (2024)

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning
by: Gehring, Jonas, et al.
Published: (2024)

The KoLMogorov Test: Compression by Code Generation
by: Yoran, Ori, et al.
Published: (2025)

BigO(Bench) -- Can LLMs Generate Code with Controlled Time and Space Complexity?
by: Chambon, Pierre, et al.
Published: (2025)

Soft Policy Optimization: Online Off-Policy RL for Sequence Models
by: Cohen, Taco, et al.
Published: (2025)

A Deep Dive into Scaling RL for Code Generation with Synthetic Data and Curricula
by: Sancaktar, Cansu, et al.
Published: (2026)

Toward Training Superintelligent Software Agents through Self-Play SWE-RL
by: Wei, Yuxiang, et al.
Published: (2025)

Meta Large Language Model Compiler: Foundation Models of Compiler Optimization
by: Cummins, Chris, et al.
Published: (2024)

The Extrapolation Power of Implicit Models
by: Decugis, Juliette, et al.
Published: (2024)

Towards a Neural Debugger for Python
by: Beck, Maximilian, et al.
Published: (2026)

Improving Diversity in Language Models: When Temperature Fails, Change the Loss
by: Verine, Alexandre, et al.
Published: (2025)

Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning
by: Hassid, Michael, et al.
Published: (2025)

CWM: An Open-Weights LLM for Research on Code Generation with World Models
by: FAIR CodeGen team, et al.
Published: (2025)

The Larger the Better? Improved LLM Code-Generation via Budget Reallocation
by: Hassid, Michael, et al.
Published: (2024)

SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution
by: Wei, Yuxiang, et al.
Published: (2025)

CodeIt: Self-Improving Language Models with Prioritized Hindsight Replay
by: Butt, Natasha, et al.
Published: (2024)

WybeCoder: Verified Imperative Code Generation
by: Gloeckle, Fabian, et al.
Published: (2026)

Extrapolation Merging: Keep Improving With Extrapolation and Merging
by: Lin, Yiguan, et al.
Published: (2025)

Beyond pass@k: Redundancy-Aware RLVR for Multi-Sample Code Generation
by: Florian, Le Bronnec, et al.
Published: (2026)

Planning without Search: Refining Frontier LLMs with Offline Goal-Conditioned RL
by: Hong, Joey, et al.
Published: (2025)

Self-Execution Simulation Improves Coding Models
by: Maimon, Gallil, et al.
Published: (2026)

Model Extrapolation Expedites Alignment
by: Zheng, Chujie, et al.
Published: (2024)

Round-Trip Translation Reveals What Frontier Multilingual Benchmarks Miss
by: Skorobogat, Ronald, et al.
Published: (2026)

WARM: On the Benefits of Weight Averaged Reward Models
by: Ramé, Alexandre, et al.
Published: (2024)

ECCO: Can We Improve Model-Generated Code Efficiency Without Sacrificing Functional Correctness?
by: Waghjale, Siddhant, et al.
Published: (2024)

ContextRL: Enhancing MLLM's Knowledge Discovery Efficiency with Context-Augmented RL
by: Lu, Xingyu, et al.
Published: (2026)

Optimizing Language Models for Inference Time Objectives using Reinforcement Learning
by: Tang, Yunhao, et al.
Published: (2025)

Mining the Mind: What 100M Beliefs Reveal About Frontier LLM Knowledge
by: Ghosh, Shrestha, et al.
Published: (2025)

Monte Carlo Graph Coloring
by: Cazenave, Tristan, et al.
Published: (2025)

CodeRL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment
by: Jiang, Xue, et al.
Published: (2025)

Incomplete Tasks Induce Shutdown Resistance in Some Frontier LLMs
by: Schlatter, Jeremy, et al.
Published: (2025)

Scaling Laws of RoPE-based Extrapolation
by: Liu, Xiaoran, et al.
Published: (2023)

Extrapolation by Association: Length Generalization Transfer in Transformers
by: Cai, Ziyang, et al.
Published: (2025)

Getting the most out of your tokenizer for pre-training and domain adaptation
by: Dagan, Gautier, et al.
Published: (2024)

Rethinking Code Refinement: Learning to Judge Code Efficiency
by: Seo, Minju, et al.
Published: (2024)

Large Language Models for Extrapolative Modeling of Manufacturing Processes
by: Khanghah, Kiarash Naghavi, et al.
Published: (2025)

LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuning
by: Perin, Gabriel J., et al.
Published: (2025)

From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debugging
by: Shi, Yuling, et al.
Published: (2024)

Generalizable End-to-End Tool-Use RL with Synthetic CodeGym
by: Du, Weihua, et al.
Published: (2025)

Scaling Test-Time Compute for Agentic Coding
by: Kim, Joongwon, et al.
Published: (2026)