PaperClaw: Harnessing Agents for Autonomous Research and Human-in-the-Loop Refinement

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Weiwei Ye, Hangchen Liu, Dongyuan Li, Renhe Jiang
• Venue: arXiv: cs.AI
• Published: 2026-06-21
• Collected: 2026-06-22
• URL: http://arxiv.org/abs/2606.22610v1
• PDF: https://arxiv.org/pdf/2606.22610v1
• DOI: UNKNOWN

Plain-Language Brief

The paper introduces PAPERCLAW, a multi-agent system that automates the research process from start to finish. It uses a large language model as its core, which can drive the entire pipeline and also allows for human intervention for refinement.

What the Document Says

• PAPERCLAW is a multi-agent system designed to carry out a project autonomously from the field of study to a finished paper.
• It curates a domain from live literature, datasets, and code, and generates an idea based on a pre-registered main-result contract.
• The system follows an iterative propose, test, reflect loop to drive a hypothesis map, stopping when evidence supports the idea.
• Once the evidence supports the idea, it writes a venue-compliant paper.
• A full-lifecycle memory keeps each stage in a single living record.
• An in-cycle research assistant with research tools and skills drives the pipeline, but allows for human intervention at any stage.
• The system ensures its output is grounded and checkable by citing only references validated against open scholarly indexes and reporting results that genuinely ran.
• An evaluation with an LLM judge found that PAPERCLAW produces strong papers both autonomously and with human-in-the-loop refinement.

Key Concepts Mentioned

• PAPERCLAW
• multi-agent system
• research automation
• large language model
• iterative propose, test, reflect loop
• human-in-the-loop refinement
• full-lifecycle memory
• scholarly indexes

Claims Made by the Document

• PAPERCLAW can carry out a project autonomously from the field of study to a finished paper.
• It produces strong papers both autonomously and with human-in-the-loop refinement.

Evidence Used

• The text describes PAPERCLAW's capabilities and an evaluation with an LLM judge.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Large language models have become capable reasoners and tool users that write and run code and search the literature, which makes automating the research process itself a realistic goal. We present PAPERCLAW, a harnessed multi-agent system that carries a project autonomously, from a field of study to a finished paper. PAPERCLAW curates a domain from a field's live literature, datasets, and code; brainstorms it into an idea with a pre-registered main-result contract; and drives a stoppable hypothesis map through an iterative propose, test, reflect loop that grows only from measured verdicts and halts once the evidence supports the idea, at which point it writes a venue-compliant paper. A full-lifecycle memory keeps each stage in a single living record, so a long run can be paused, inspected, and resumed without losing context. At the centre is an in-cycle research assistant with research tools and skills: it can drive the whole pipeline on its own, while the same interface lets a person step in at any stage, turning a first autonomous draft into a stronger paper through human-in-the-loop refinement. Throughout, PAPERCLAW keeps its output grounded and checkable, citing only references validated against open scholarly indexes and reporting results that genuinely ran. An evaluation with an LLM judge finds that PAPERCLAW produces strong papers both fully autonomously and with human-in-the-loop refinement.