Paper page - AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery

Source: https://huggingface.co/papers/2605.23204 Published on May 22

·

Submitted byhttps://huggingface.co/tgy2024

2024on May 26

#3 Paper of the day Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

AI systems are evolving from task-specific assistants to workflow-level research automators, facing challenges in autonomy, reproducibility, and accountability across scientific domains.

Scientific research is being reshaped by AI systems that move beyond isolated assistance toward longer-horizon workflows spanning literature grounding, hypothesis generation, experimentation, validation, reporting, and revision. This shift marks a transition from task-level AI for science to workflow-level research automation. Yet current systems remain fragmented, differing in autonomy, domain scope, execution environment, validation mechanism, and human oversight, while still struggling with evidence preservation, reproducibility, weak-direction rejection,provenancetracking, cross-domain robustness, and accountable scientific closure. This survey examines these developments throughAutoResearch, defined as the developmental spectrum ofAI-powered scientific workflow automation. Within it,Vibe Researchdenotes the human-steered region of prompt-based assistance and human-verified execution, whereas emerging AI-led systems coordinate larger portions of the discovery loop without achieving robust autonomy. We analyze how research systems redistribute control, evidence, execution, validation, and accountability across workflows and organize the field around five workflow conditions: literature and research grounding; hypothesis formation and planning; experimentation and tool use; feedback, validation, and review; and reporting and knowledge communication. We further synthesize AI scientist systems,mixed-initiative co-research frameworks,benchmarks,domain deployments, andopen-source infrastructures. Finally, we propose fiveevaluation dimensions--novelty,validity,impact,reliability, andprovenance--and show thatAutoResearchautonomy is domain-conditioned, being more credible in structured, executable, and rapidly verifiable settings but limited in embodied, delayed, heterogeneous, ethical, or institutionally accountable contexts.

View arXiv pageView PDFProject pageGitHub0Add to collection

Get this paper in your agent:

hf papers read 2605\.23204

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.23204 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.23204 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.23204 in a Space README.md to link it from this page.

Collections including this paper1