mlsanityData to 14 Oct 2026

Ockhamareto: Pareto-Gated Segment-Level Credit Assignment for Concise Unit-Test Generation with Reinforcement Learning

arXiv 2608.24473 ↗First posted 25 Aug 2026cs.SE cs.CLcited by 1

On the map: LLM reasoning → Reinforcement learning for reasoning

also in 1 more
Abstract (arXiv)

We introduce Ockhamareto, a single-shot GRPO framework for unit-test generation and selection, based on the principles of Ockham's Razor and Pareto Optimality. Ockhamareto has two principal components: (i)~a Pareto-gated Bonus that rewards only rollouts non-dominated in~(mutation, -#tests) space, and (ii)~Token-level Segment Credit, which attributes each test's marginal mutation kills back to the tokens of its unit-test block. On the UnLeakedTestBench~(ULT), Ockhamareto strictly Pareto-dominates the strongest RL baseline~(MIST-RL). Furthermore, it dominates on em each and all optimization objectives, catching more bugs (49.9% vs 31.3% mutation score at N=5), using fewer tests (2.60 vs 4.67 on average), thereby achieving 3.4× the per-test trade-off improvement. The advantage is found in all four benchmarks~(HumanEval+, MBPP+, CodeContests, TestGenEval-Lite): Ockhamareto leads both mutation and coverage metrics on every one, always with the smallest suite. Ockhamareto also outperforms the state-of-the-art at all model scales, adding +30--35~pp mutation at 4B, 9B, and 27B model sizes. We also show that the knee point of the optimal trade-off between efficiency and effectiveness on the Pareto front is not correlated with obvious more easily computed proxy metrics, such as function size. This finding motivates the Pareto front computation; it is needed to identify this crucial engineering trade-off for each function under test.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning. Builds on 16 of its 34 references that are in the map. First posted on arXiv 25 Aug 2026.

Builds on

16 of 34 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023Let's Verify Step by Step31 May
  2. 2017Proximal Policy Optimization Algorithms20 Jul

Other references · 14

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  5. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,373
  6. 2022Competition-Level Code Generation with AlphaCodecited by 1,462
  7. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,391
  8. 2023An Empirical Evaluation of Using Large Language Models for Automated Unit Test Generationcited by 194
  9. 2023No More Manual Tests? Evaluating and Improving ChatGPT for Unit Test Generationcited by 82
  10. 2024Code-Aware Prompting: A study of Coverage Guided Test Generation in Regression Setting using LLMcited by 64
  11. 2023Using Large Language Models to Generate JUnit Tests: An Empirical Studycited by 53
  12. 2024TestGenEval: A Real World Unit Test Generation and Test Completion Benchmarkcited by 52
2 more
  1. 2022Pynguin: Automated Unit Test Generation for Pythoncited by 50
  2. 2025Benchmarking LLMs for Unit Test Generation from Real-World Functionscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Teaching Agents to Code Reliablycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026MIST-RL: Mutation-based Incremental Suite Testing via Reinforcement Learning
  2. 2026Blockwise Advantage Estimation for Multi-Objective RL with Verifiable Rewards
  3. 2025Co-Evolving LLM Coder and Unit Tester via Reinforcement Learning
  4. 2025MURPHY: Feedback-Aware GRPO with Retrospective Credit Assignment for Multi-Turn Code Generation
  5. 2026Code-A1: Adversarial Evolving of Code LLM and Test LLM via Reinforcement Learning
  6. 2025Segment Policy Optimization: Effective Segment-Level Credit Assignment in RL for Large Language Models
  7. 2026Entropy-Regularized Rank-Masked Policy Optimization for Test-Time Reinforcement Learning in Code Generation
  8. 2026Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences