Pith. sign in

REVIEW 8 cited by

ACECODER: Acing Coder RL via Automated Test-Case Synthesis

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.01718 v4 pith:XIHZVD46 submitted 2025-02-03 cs.SE cs.AIcs.CL

ACECODER: Acing Coder RL via Automated Test-Case Synthesis

Huaye Zeng , Dongfu Jiang , Haozhe Wang , Ping Nie , Xiaotong Chen , Wenhu Chen This is my paper
classification cs.SE cs.AIcs.CL
keywords modelmodelscodecoderlearningreinforcementrewardtest-case
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Most progress in recent coder models has been driven by supervised fine-tuning (SFT), while the potential of reinforcement learning (RL) remains largely unexplored, primarily due to the lack of reliable reward data/model in the code domain. In this paper, we address this challenge by leveraging automated large-scale test-case synthesis to enhance code model training. Specifically, we design a pipeline that generates extensive (question, test-cases) pairs from existing code data. Using these test cases, we construct preference pairs based on pass rates over sampled programs to train reward models with Bradley-Terry loss. It shows an average of 10-point improvement for Llama-3.1-8B-Ins and 5-point improvement for Qwen2.5-Coder-7B-Ins through best-of-32 sampling, making the 7B model on par with 236B DeepSeek-V2.5. Furthermore, we conduct reinforcement learning with both reward models and test-case pass rewards, leading to consistent improvements across HumanEval, MBPP, BigCodeBench, and LiveCodeBench (V4). Notably, we follow the R1-style training to start from Qwen2.5-Coder-base directly and show that our RL training can improve model on HumanEval-plus by over 25\% and MBPP-plus by 6\% for merely 80 optimization steps. We believe our results highlight the huge potential of reinforcement learning in coder models.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models

    cs.CL 2026-01 conditional novelty 7.0

    Arbitrary-order generation lowers the reasoning Pass@k of diffusion LLMs; constraining RL to autoregressive order (JustGRPO) improves accuracy and preserves parallel decoding.

  2. SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution

    cs.SE 2025-02 unverdicted novelty 7.0

    SWE-RL uses RL on software evolution data to train LLMs achieving 41% on SWE-bench Verified with generalization to other reasoning tasks.

  3. GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language Models

    cs.LG 2026-05 unverdicted novelty 6.0

    GDSD reduces RL for dLLMs to likelihood-free self-distillation via a normalization-free logit-matching objective, outperforming ELBO methods with more stable training on LLaDA-8B and Dream-7B.

  4. Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts

    cs.LG 2026-04 unverdicted novelty 6.0

    BAR trains independent domain experts via separate mid-training, SFT, and RL pipelines then composes them with a MoE router to match monolithic retraining performance at lower cost and without catastrophic forgetting.

  5. ZeroCoder: Can LLMs Improve Code Generation Without Ground-Truth Supervision?

    cs.SE 2026-04 unverdicted novelty 6.0

    Co-evolving coder and tester models via consensus over a self-generated pass matrix improves LLM code generation up to 14.5% label-free and 21.6% with a lightly calibrated Bayesian selector.

  6. ZeroCoder: Can LLMs Improve Code Generation Without Ground-Truth Supervision?

    cs.SE 2026-04 unverdicted novelty 6.0

    ZeroCoder co-evolves coder and tester LLMs via self-generated code-test execution feedback to improve code generation up to 21.6% without ground-truth supervision.

  7. InfoSynth: Information-Guided Benchmark Synthesis for LLMs

    cs.CL 2026-01 conditional novelty 5.0

    Using KL divergence and entropy on embeddings, InfoSynth scores benchmark novelty/diversity and guides a genetic pipeline that generates new, code-verified Python problems from seeds.

  8. CoTu at EXACT 2026: Neuro-Symbolic Reasoning for Transparent Educational QA

    cs.CL 2026-07 conditional novelty 4.0

    A 4B LLM that generates solver-runnable code (Z3 for logic, Python for physics) achieved a perfect physics score and the highest technical score (13.44/15) at EXACT 2026, with premise selection as the main remaining e...