# ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents

arXiv: https://arxiv.org/abs/2609.17523 · HTML: https://mlsanity.com/paper/2609.17523/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 28 references that are in the map.
- First posted on arXiv 15 Sep 2026.

## Builds on (most cited first, 25 of 25)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models](https://mlsanity.com/paper/2510.04618/) (2025)
- [Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains](https://mlsanity.com/paper/2507.17746/) (2025)
- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [Meta-Harness: End-to-End Optimization of Model Harnesses](https://mlsanity.com/paper/2603.28052/) (2026)
- [LAB-Bench: Measuring Capabilities of Language Models for Biology Research](https://mlsanity.com/paper/2407.10362/) (2024)
- [SkillOpt: Executive Strategy for Self-Evolving Agent Skills](https://mlsanity.com/paper/2605.23904/) (2026)
- [OpenClaw-RL: Train Any Agent Simply by Talking](https://mlsanity.com/paper/2603.10165/) (2026)
- [Self-Adapting Language Models](https://mlsanity.com/paper/2506.10943/) (2025)
- [AgentRx: Diagnosing AI Agent Failures from Execution Trajectories](https://mlsanity.com/paper/2602.02475/) (2026)
- [Hyperagents](https://mlsanity.com/paper/2603.19461/) (2026)
- [SkillGen: Verified Inference-Time Agent Skill Synthesis](https://mlsanity.com/paper/2605.10999/) (2026)
- [RLAnything: Forge Environment, Policy, and Reward Model in Completely Dynamic RL System](https://mlsanity.com/paper/2602.02488/) (2026)
- [SIA: Self Improving AI with Harness & Weight Updates](https://mlsanity.com/paper/2605.27276/) (2026)
- [AdaRubric: Task-Adaptive Rubrics for Reliable LLM Agent Evaluation and Reward Learning](https://mlsanity.com/paper/2603.21362/) (2026)
- [HELIX: Model-Harness Co-evolution for Recursive Self-Improvement](https://mlsanity.com/paper/2608.13951/) (2026)
- [Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programs](https://mlsanity.com/paper/2604.06699/) (2026)
- [Reinforcement Learning with Robust Rubric Rewards](https://mlsanity.com/paper/2605.30244/) (2026)
- [PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents](https://mlsanity.com/paper/2608.26530/) (2026)
- [Fantastic Scientific Agents and How to Build Them: AgentBuild for Rietveld Refinement](https://mlsanity.com/paper/2606.12834/) (2026)
- [Environment Evolution for Terminal Agents](https://mlsanity.com/paper/2609.04128/) (2026)

## Built on by (most cited first, 7 of 7)

- [ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments](https://mlsanity.com/paper/2609.19134/) (2026)
- [JEPA-Anything: Learning Predictive Models across Different Worlds](https://mlsanity.com/paper/2609.20800/) (2026)
- [Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite](https://mlsanity.com/paper/2610.02826/) (2026)
- [Intervention anchors and scientific verification in synthetic vascular predictive representations](https://mlsanity.com/paper/2610.11704/) (2026)
- [RSI-Anything: Unlocking Recursive Self-Improvement with World Models](https://mlsanity.com/paper/2610.13912/) (2026)
- [ScienceBuddy-PV: A Verifier-Governed Recursive Improvement Protocol for Panvascular Intervention Research](https://mlsanity.com/paper/2610.14523/) (2026)
- [ECG-RSI: Recursive Self-Improvement through Targeted Electrocardiogram Waveform Evolution](https://mlsanity.com/paper/2610.17718/) (2026)

## Abstract (arXiv)

We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers' everyday workflows. ScienceBuddy supports researchers in carrying out scientific tasks while transforming their requests, feedback, and execution evidence into tasks and evaluation rubrics for continual learning. At its core is recursive-in-recursive self-improvement, a paradigm that couples harness evolution with model reinforcement learning: the inner recursion improves the harness with the model fixed, while the outer recursion trains the model under the improved harness. Harness evolution shapes training experience, and model learning creates new opportunities for harness adaptation. We present case studies of researcher interaction, harness refinement, and model learning, with the benchmark cases spanning four scientific task families. By releasing ScienceBuddy as a research product, we make this paradigm available to the scientific community and take a step toward discovery intelligence: scientific AI that advances through sustained collaboration with researchers and evolves alongside the research it supports. Website: http://science-buddy.io