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

Hugging Face Daily Papers Papers

Summary

ScienceBuddy introduces a recursive-in-recursive self-improvement paradigm for interactive scientific agents, enabling continual evolution through researcher collaboration and feedback.

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
Original Article
View Cached Full Text

Cached at: 09/16/26, 02:45 AM

Paper page - ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents

Source: https://huggingface.co/papers/2609.17523 Published on Sep 15

#2 Paper of the day Authors:

,

,

,

,

,

,

,

,

,

,

,

Abstract

WeintroduceandreleaseScienceBuddy,aninteractivescientificresearchworkspacethatbringscontinuallyimprovingscientificagentsintoresearchers’everydayworkflows.ScienceBuddysupportsresearchersincarryingoutscientifictaskswhiletransformingtheirrequests,feedback,andexecutionevidenceintotasksandevaluationrubricsforcontinuallearning.Atitscoreisrecursive-in-recursiveself-improvement,aparadigmthatcouplesharnessevolutionwithmodelreinforcementlearning:theinnerrecursionimprovestheharnesswiththemodelfixed,whiletheouterrecursiontrainsthemodelundertheimprovedharness.Harnessevolutionshapestrainingexperience,andmodellearningcreatesnewopportunitiesforharnessadaptation.Wepresentcasestudiesofresearcherinteraction,harnessrefinement,andmodellearning,withthebenchmarkcasesspanningfourscientifictaskfamilies.ByreleasingScienceBuddyasaresearchproduct,wemakethisparadigmavailabletothescientificcommunityandtakeasteptowarddiscoveryintelligence:scientificAIthatadvancesthroughsustainedcollaborationwithresearchersandevolvesalongsidetheresearchitsupports.Website:http://science-buddy.io

View arXiv pageView PDFProject pageGitHub5Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.17523 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.17523 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.17523 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Recursive self-improvement of AI research agents

Hugging Face Daily Papers

This paper introduces AIDE^2, a system that enables AI research agents to autonomously improve their own code through recursive self-improvement, leading to performance gains across various AI research tasks.

The Economics of Recursive Self-Improvement [pdf]

Hacker News Top

This paper examines the economic incentives and dynamics of recursive self-improvement in AI systems, addressing how such processes could scale and their implications for governance and safety.

AREX: Towards a Recursively Self-Improving Agent for Deep Research

Hugging Face Daily Papers

AREX introduces a family of recursively self-improving agents for deep research, alternating between an inner research loop and an outer self-improvement loop, trained with long-horizon reinforcement learning. It substantially outperforms comparable-scale baselines on benchmarks like BrowseComp and Humanity's Last Exam.