# Beneath the Diff: Diagnosing and Mitigating Algorithmic Mode Collapse in Code-Level Autonomous Research Loops

arXiv: https://arxiv.org/abs/2609.00077 · HTML: https://mlsanity.com/paper/2609.00077/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 50 references that are in the map.
- First posted on arXiv 31 Aug 2026.

## Builds on (most cited first, 21 of 21)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge](https://mlsanity.com/paper/1811.00937/) (2018)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery](https://mlsanity.com/paper/2408.06292/) (2024)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Are NLP Models really able to Solve Simple Math Word Problems?](https://mlsanity.com/paper/2103.07191/) (2021)
- [Agent Laboratory: Using LLM Agents as Research Assistants](https://mlsanity.com/paper/2501.04227/) (2025)
- [The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models](https://mlsanity.com/paper/2201.03544/) (2022)
- [A Note on Shumailov et al. (2024): `AI Models Collapse When Trained on Recursively Generated Data'](https://mlsanity.com/paper/2410.12954/) (2024)
- [All That Glitters is Not Novel: Plagiarism in AI Generated Research](https://mlsanity.com/paper/2502.16487/) (2025)
- [Why LLMs Aren't Scientists Yet: Lessons from Four Autonomous Research Attempts](https://mlsanity.com/paper/2601.03315/) (2026)
- [R-Diverse: Mitigating Diversity Illusion in Self-Play LLM Training](https://mlsanity.com/paper/2602.13103/) (2026)
- [Recursive Training Loops in LLMs: How training data properties modulate distribution shift in generated data?](https://mlsanity.com/paper/2504.03814/) (2025)
- [Preventing Curriculum Collapse in Self-Evolving Reasoning Systems](https://mlsanity.com/paper/2603.13309/) (2026)

## Built on by (most cited first, 1 of 1)

- [Scaling Discovery Requires Forgetting](https://mlsanity.com/paper/2610.20563/) (2026)

## Abstract (arXiv)

Code-level autonomous research loops (ARLs) have recently emerged as a concrete object of study in automated machine learning research. In such loops, an LLM agent proposes modifications to an experimental training pipeline, executes the modified pipeline, and retains edits that improve a verifiable in-loop metric. Although executable metrics may appear to provide a reliable signal of progress, it remains unclear whether repeated metric-driven code editing leads to genuine improvements that generalize beyond the loop. We provide a systematic diagnosis of this question. Across various experiment settings, we identify a robust failure mode that we call algorithmic mode collapse. In this regime, surface-level edit diversity remains stable, but semantic and mechanism-level diversity collapse: the agent continues to edit different lines of code while repeatedly proposing the same kinds of algorithmic changes. This collapse is accompanied by a widening gap between in-loop metric gains and gains measured on independent held-out evaluations. We then propose Diversity-Aware Proposal Sampling (textscDAPS), a lightweight mitigation that combines category-coverage reweighting, persistent edit memory, and a validation gate. Under a three-tier protocol separating the in-loop metric, the audit metric read by the gate, and a blind metric no loop component ever accesses, textscDAPS reduces semantic-cluster decay of edits by 69.1% and improves relative faithfulness by 83.7% blind and 81.6% audited, while preserving in-loop optimization speed. We provide the code in Github hrefhttps://github.com/BokwaiHo/arl-mode-collapserepository.