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

PAPERCLAW is a multi-agent system designed to automate the research process from literature curation to paper writing. It operates through an iterative propose-test-reflect loop, utilizing a full-lifecycle memory to maintain context, allowing for pausing and resuming projects. The system integrates a human-in-the-loop mechanism for refinement and has been evaluated against an LLM judge, demonstrating its capability to produce high-quality research papers autonomously and with human input, which is significant for practitioners aiming to enhance research efficiency and output quality in AI.