BoostAPR: Boosting Automated Program Repair via Execution-Grounded Reinforcement Learning with Dual Reward Models

• URL: http://arxiv.org/abs/2605.09134v3
• Date: Wed, 13 May 2026 16:38:24 GMT
• Title: BoostAPR: Boosting Automated Program Repair via Execution-Grounded Reinforcement Learning with Dual Reward Models
• Abstract summary: Reinforcement learning for program repair is hindered by sparse execution feedback and coarse sequence-level rewards that obscure which edits actually fix bugs.<n>We present BoostAPR, a three-stage framework addressing these challenges.
• Score: 9.519307176717094
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning for program repair is hindered by sparse execution feedback and coarse sequence-level rewards that obscure which edits actually fix bugs. We present BoostAPR, a three-stage framework addressing these challenges: (1) supervised fine-tuning on execution-verified demonstrations with reasoning traces, (2) training dual reward models--a sequence-level assessor and a line-level credit allocator--from execution outcomes, and (3) PPO optimization where the line-level model redistributes rewards to critical edit regions. This line-level credit assignment operates at an intermediate granularity naturally suited to code changes. Trained on SWE-Gym and evaluated on four benchmarks, BoostAPR achieves 40.7% on SWE-bench Verified (+22.9pp over base model), 24.8% on Defects4J (Python-to-Java transfer), 84.5% on HumanEval-Java, and 95.0% on QuixBugs, achieving competitive results among open-source models with strong cross-language generalization.

Related papers

• Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization [60.89431018071735]
  Reinforcement learning with group-relative advantages has become the de facto standard for post-training language model reasoners.<n>We show that rollouts with distinct reward profiles can receive identical advantages, and all objectives are optimized with fixed relative weights regardless of their current level of saturation.<n>We introduce saturation-aware reweighting, which standardizes each reward objective independently and adaptively discounts its contribution according to a batch-level estimate of objective saturation.
  arXiv  Detail & Related papers  (2026-08-17T04:07:50Z)
• TaPR: Test-Aware Policy Refinement for Feedback-Conditioned Code Generation [6.268849049716121]
  Test-aware Policy Refinement (TaPR) is a framework that transforms execution feedback into a dense per-turn test-pass-ratio reward.<n>Across six models on 219 code-generation problems from LiveCodeBench, TaPR improves the pooled three-turn success rate (Pass@3) by 2.44 percentage points.
  arXiv  Detail & Related papers  (2026-08-01T07:35:06Z)
• SCOPE: Leveraging Subgoal Critiques for Code Generation [14.869703886733761]
  We present SCOPE, a prover-d subgoal critic for code generation.<n>SCOPE adapts a Lean-oriented prover model to produce three parseable feedback fields for downstream code generation.<n>Our approach combines supervised fine-tuning, process-aligned reinforcement learning (RL), and feedback-guided inference.
  arXiv  Detail & Related papers  (2026-07-07T04:09:41Z)
• ExTra: Exploratory Trajectory Optimization for Language Model Reinforcement Learning [75.26064451955199]
  ExTra is a framework that extracts exploration signals from the model's own rollouts.<n>It improves Qwen3-1.7B over GRPO by about +5 points on pass@1 and +7 points on pass@16.
  arXiv  Detail & Related papers  (2026-06-23T15:51:39Z)
• VeriGate: Verifier-Gated Step-Level Supervision for GRPO [51.26100506256885]
  Group Relative Policy Optimization is an effective recipe for training reasoning models with verifier-based outcome rewards.<n>We present VeriGate, a verifier-gated extension of GRPO that addresses these limitations with three design choices.<n>We show that VeriGate improves average accuracy by about 20% and 12% for 1.5B and 7B models respectively.
  arXiv  Detail & Related papers  (2026-05-28T18:20:32Z)
• Converted, Not Equivalent: Benchmarking Codebase Conversion via Observational Equivalence [56.25095230687242]
  Coding agents often over-trust their own local validation routines and declare success on artifacts that satisfy surface checks.<n>This problem is especially acute in conversion, where prior evaluation is outcome-driven.<n>Across blind conversion attempts, the best system reaches 26.7--28.9% overall pass rate despite Spec pass rates up to 91.1%.<n>This suggests that failures stem more from contract-misaligned self-validation than from limited budget or backbone strength.
  arXiv  Detail & Related papers  (2026-05-27T19:57:15Z)
• Step-wise Rubric Rewards for LLM Reasoning [72.17879367869503]
  Reinforcement Learning with Verifiable Rewards (RLVR) is widely used to improve reasoning in large language models.<n>We find 18.2% of steps in correct responses are wrong yet positively rewarded.<n>Across six mathematical reasoning benchmarks, SRaR improves average accuracy over RaR by 3.57 points.
  arXiv  Detail & Related papers  (2026-05-17T07:08:14Z)
• Multi-Turn Reinforcement Learning for Tool-Calling Agents with Iterative Reward Calibration [0.0]
  Training tool-calling agents with reinforcement learning on multi-turn tasks remains challenging.<n>We present the first application of MT-GRPO and GTPO for training a tool-calling agent on realistic customer service tasks.
  arXiv  Detail & Related papers  (2026-04-03T08:36:03Z)
• On the Effectiveness of Code Representation in Deep Learning-Based Automated Patch Correctness Assessment [27.074607600022315]
  Automated program repair (APR) attempts to generate correct patches and has drawn wide attention from both academia and industry in the past decades.<n>To address the overfitting problem, the community has proposed an increasing number of approaches to predict patch correctness ( APCA approaches)<n>Among them, locally deep learning approaches aimed at automatically match designs has been emerging strongly.<n>Despite being fundamental in reason about patch correctness, code representation has not been systematically investigated.
  arXiv  Detail & Related papers  (2026-03-08T08:18:42Z)
• SPARK: Stepwise Process-Aware Rewards for Reference-Free Reinforcement Learning [39.1720897614261]
  Process reward models (PRMs) that provide dense, step-level feedback have shown promise for reinforcement learning.<n>We propose SPARK: a three-stage framework where in the first stage a generator model produces diverse solutions and a verifier model evaluates them.<n>We show that aggregating multiple independent verifications at the step level produces training data for process reward models that surpass ground-truth outcome supervision.
  arXiv  Detail & Related papers  (2025-12-02T21:30:47Z)
• Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric Domains [97.5573252172065]
  We train a family of Automatic Reasoning Evaluators (FARE) with a simple iterative rejection-sampling supervised finetuning approach.<n>FARE-8B challenges larger specialized RL-trained evaluators and FARE-20B sets the new standard for open-source evaluators.<n>As inference-time rerankers, FARE-20B achieves near-oracle performance on MATH.
  arXiv  Detail & Related papers  (2025-10-20T17:52:06Z)
• Automated Repair of C Programs Using Large Language Models [0.0]
  This study explores the potential of Large Language Models (LLMs) in automating the repair of C programs.<n>We present a framework that integrates spectrum-based fault localization (SBFL), runtime feedback, and Chain-of-Thought-structured prompting into an autonomous repair loop.<n>Our approach achieves 44.93% repair accuracy, representing a 3.61% absolute improvement over strong state-of-the-art APR baselines.
  arXiv  Detail & Related papers  (2025-09-02T04:34:11Z)
• Critique-GRPO: Advancing LLM Reasoning with Natural Language and Numerical Feedback [59.078756231841574]
  Critique-GRPO is an online RL framework that integrates both natural language and numerical feedback for effective policy optimization.<n>We show Critique-GRPO consistently outperforms supervised learning and RL-based fine-tuning methods across eight challenging mathematical, STEM, and general reasoning tasks.
  arXiv  Detail & Related papers  (2025-06-03T17:39:02Z)
• The Art of Repair: Optimizing Iterative Program Repair with Instruction-Tuned Models [48.073219761367184]
  We investigate an APR pipeline that balances the generation of multiple outputs and multiple rounds of iteration.<n>We fine-tune each model on an APR dataset with three sizes (1K, 30K, 65K) and two techniques (Full Fine-Tuning and LoRA)<n>Our results show that by using only a fraction (1%) of the fine-tuning dataset, we can achieve improvements of up to 78% in the number of plausible patches generated.
  arXiv  Detail & Related papers  (2025-05-05T18:06:51Z)
• EquiBench: Benchmarking Large Language Models' Reasoning about Program Semantics via Equivalence Checking [58.15568681219339]
  We introduce EquiBench, a new benchmark for evaluating large language models (LLMs)<n>This task directly tests a model's ability to reason about program semantics.<n>We evaluate 19 state-of-the-art LLMs and find that in the most challenging categories, the best accuracies are 63.8% and 76.2%, only modestly above the 50% random baseline.
  arXiv  Detail & Related papers  (2025-02-18T02:54:25Z)
• Reinforcement Learning for Mutation Operator Selection in Automated Program Repair [11.756822700775668]
  We investigate the feasibility of a reinforcement learning-based approach for the selection of mutation operators in program repair. Our proposed approach is language, programming-level, and search strategy and allows for easy augmentation into existing-based repair tools. We evaluate our approach on 353 real-world bugs from the Defects4J benchmark.
  arXiv  Detail & Related papers  (2023-06-09T10:09:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.