Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

Hugging Face Research: RL Boosts Code Optimization Accuracy by Up to 125%

AI By Crimson AI Hugging Face Papers 29 July 2026 · 00:00 12 views
Share: X Telegram

A new Hugging Face paper introduces DMC-Optim and a three-stage RL framework that turns execution time into a learnable signal, improving top-50% pass@1 by up to 19.7 points on Qwen 2.5 7B and CWM 32B while preserving correctness.

Hugging Face Research: RL Boosts Code Optimization Accuracy by Up to 125%

Key points

Reinforcement learning (RL) has proven effective for code correctness, where models generate programs and are rewarded for passing hidden tests. Extending this to code optimization—rewarding faster execution—has been challenging due to measurement noise, reward sparsity, and instability in GRPO, often leading to marginal speed gains and more failures.

In a new paper, Hugging Face researchers propose a three-stage approach to make execution time learnable. First, they build DMC-Optim, a benchmark with large optimization tests and a calibrated sandbox for reliable timing. Second, they design a reward function that combines correctness and speed, using an offline simulator to predict the most promising configurations. Third, they adapt GRPO and evaluation to handle the sparser, noisier timed-execution setting.

Results on DMC-Optim show significant improvements: strict top-50% pass@1 rises from 18.0% to 31.3% on Qwen 2.5 7B and from 30.7% to 50.4% on CWM 32B. At stricter percentiles like top-30%, CWM 32B achieves a 125% relative improvement while preserving pure-correctness scores. When the timing sandbox is degraded, robust optimization RL outperforms standard RLVR by 100% to 200% depending on the evaluation criterion.

On the LCB benchmark, CWM 32B wins up to 83% of median-sample speed comparisons against standard RLVR. Relative to the fastest correct human submissions per problem, it achieves about half the human rate of complexity-class improvements (14% vs. 28%).

ModelMetricStandard RLVROptimization-Aware RL
Qwen 2.5 7BTop-50% pass@118.0%31.3%
CWM 32BTop-50% pass@130.7%50.4%
CWM 32BTop-30% relative improvement+125%
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 31 Aug 2026

Hugging Face Unveils StepGuard: Step-Level Guardrails for Safer AI Agents

StepGuard, a new step-level guard model from Hugging Face, audits agent actions before execution, reducing attack success rates by...

1
Research paper
Hugging Face 31 Aug 2026

Hugging Face Researchers Unveil ABot-Recon for Stable Long-Horizon 3D Reconstruction

ABot-Recon, a new streaming 3D reconstruction model from Hugging Face, achieves stable long-horizon performance using only local t...

1
Research paper
Hugging Face 31 Aug 2026

ContextPilot: Teaching Agents Proactive Context Management via Fine-Grained RL

Hugging Face researchers introduce ContextPilot, a framework that enhances long-horizon agent reasoning by expanding context-editi...

1