# Don't Repeat Yourself: Self-Supervised Fine-Tuning for Coverage

arXiv: https://arxiv.org/abs/2609.31688 · HTML: https://mlsanity.com/paper/2609.31688/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 38 references that are in the map.
- First posted on arXiv 16 Sep 2026.

## Builds on (most cited first, 30 of 32)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Competition-Level Code Generation with AlphaCode](https://mlsanity.com/paper/2203.07814/) (2022)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?](https://mlsanity.com/paper/2504.13837/) (2025)
- [STaR: Bootstrapping Reasoning With Reasoning](https://mlsanity.com/paper/2203.14465/) (2022)
- [DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation](https://mlsanity.com/paper/2211.11501/) (2022)
- [Understanding the Effects of RLHF on LLM Generalisation and Diversity](https://mlsanity.com/paper/2310.06452/) (2023)
- [Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models](https://mlsanity.com/paper/2312.06585/) (2023)
- [2 OLMo 2 Furious](https://mlsanity.com/paper/2501.00656/) (2024)
- [Reasoning with Exploration: An Entropy Perspective](https://mlsanity.com/paper/2506.14758/) (2025)
- [Distilling System 2 into System 1](https://mlsanity.com/paper/2407.06023/) (2024)
- [Is Model Collapse Inevitable? Breaking the Curse of Recursion by Accumulating Real and Synthetic Data](https://mlsanity.com/paper/2404.01413/) (2024)
- [Learning by Distilling Context](https://mlsanity.com/paper/2209.15189/) (2022)
- [Pass@k Training for Adaptively Balancing Exploration and Exploitation of Large Reasoning Models](https://mlsanity.com/paper/2508.10751/) (2025)
- [Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity](https://mlsanity.com/paper/2510.01171/) (2025)
- [EXAONE 3.5: Series of Large Language Models for Real-world Use Cases](https://mlsanity.com/paper/2412.04862/) (2024)
- [Diverse Preference Optimization](https://mlsanity.com/paper/2501.18101/) (2025)
- [On the Effect of Sampling Diversity in Scaling LLM Inference](https://mlsanity.com/paper/2502.11027/) (2025)
- [SED-SFT: Selectively Encouraging Diversity in Supervised Fine-Tuning](https://mlsanity.com/paper/2602.07464/) (2026)
- [Training Large Language Models To Reason In Parallel With Global Forking Tokens](https://mlsanity.com/paper/2510.05132/) (2025)
- [Diversity in Large Language Models under Supervised Fine-Tuning](https://mlsanity.com/paper/2605.00195/) (2026)
- [Why Do Reasoning Models Lose Coverage? The Role of Data and Forks in the Road](https://mlsanity.com/paper/2605.17026/) (2026)
- [GiFT: Gibbs Fine-Tuning for Code Generation](https://mlsanity.com/paper/2502.11466/) (2025)
- [The Road Less Traveled: Enhancing Exploration in LLMs via Sequential Sampling](https://mlsanity.com/paper/2510.15502/) (2025)

## Abstract (arXiv)

In verifiable domains such as math and coding, finding one correct solution among many attempts can matter more than the pass rate of each attempt. Post-training can concentrate large language model outputs around a few modes, while increasing sampling temperature has limited effectiveness. We introduce Don't Repeat Yourself Supervised Fine-Tuning (DRY-SFT), a post-training method that increases output diversity and coverage: the probability of at least one correct solution among many attempts. DRY-SFT has two stages. First, for each problem, sequentially generate K solutions, showing the model all prior attempts and asking for a different solution. Second, fine-tune on each attempt independently, removing prior attempts from the context. The process uses no reward, verifier, or correctness filter. On HumanEval+, MBPP+, and DS-1000, DRY-SFT raises pass@100 by 10.8, 12.5, and 12.4 percentage points, respectively, at a small cost to pass@1. Structural diversity, measured by abstract syntax tree edit distance among passing solutions, rises significantly on all three benchmarks. DRY-SFT also solves 244 of 600 problems that the base model did not solve in the same 200 attempts. Across nine open-weight models, lower structural diversity of the base model significantly predicts larger DRY-SFT gains, indicating that the method is especially effective on more mode-collapsed models.