@AlphaSignalAI: Karpathy automated experiments. AutoResearchClaw automated the whole lab. Most AI research tools handle one step. This …
Summary
AutoResearchClaw is a GitHub repository that automates the entire AI research pipeline from an idea to a full conference paper with real experiments, verified citations, and working code, outperforming previous autonomous research systems by 54.7% on a 55-topic benchmark.
View Cached Full Text
Cached at: 05/22/26, 09:58 PM
Karpathy automated experiments. AutoResearchClaw automated the whole lab.
Most AI research tools handle one step. This one is a GitHub repo that handles all of them.
AutoResearchClaw takes one idea as input. It outputs a full conference paper with real experiments, verified citations, and working code.
Here’s what happens in between:
- Scans 50+ papers automatically
- Three agents debate the best hypothesis
- Writes and self-debugs experiment code
- Rewrites failed hypotheses from scratch
- Drafts the paper, verifies every citation
The agents aren’t generic. Specialized versions plug into real domain tools for physics, biology, and more.
To evaluate this, the team built a benchmark across 55 topics in ML, physics, and biology. On it, the repo outperforms the previous best autonomous research system by 54.7%.
Check it out this weekend.
Huaxiu Yao (@HuaxiuYaoML): 🔥 AutoResearchClaw tech report + v0.5.0 just dropped.
12,300+⭐ on GitHub. Two big additions this release:
🧪 1/ Domain-Expert Agents in the experiment stage: Specialized agents for high-energy physics, biology, and more. Real domain tools + knowledge plugged in — not a
Similar Articles
@rohanpaul_ai: New Meta, Stanford, Google and many other top labs paper proposes AutoResearchClaw. Shows that automated research impro…
A new paper from Meta, Stanford, and Google introduces AutoResearchClaw, which improves automated research by integrating failure recovery, debate, and selective human input. It outperforms AI Scientist v2 by 54.7% on ARC-Bench and reveals that autonomy is enhanced when constrained by process rather than given unlimited freedom.
AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration
AutoResearchClaw is a multi-agent autonomous research system that improves scientific discovery through structured debate, self-healing execution, and human collaboration, outperforming previous systems on the ARC-Bench benchmark by 54.7%.
AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery
A survey paper examining the transition of AI from task-specific assistants to workflow-level research automators, defining AutoResearch as the spectrum of AI-powered scientific workflow automation and analyzing challenges in autonomy, reproducibility, and accountability.
AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery
This survey examines the emerging field of AI-powered research automation (AutoResearch), analyzing how AI systems are moving from isolated task assistance to full workflow-level scientific discovery. It defines a spectrum from human-steered 'Vibe Research' to AI-led systems, and proposes five evaluation dimensions for scientific credibility.
@GitHub_Daily: AI guru Karpathy previously built an autoresearch script that lets the model automatically run hundreds of rounds of experimental tuning overnight. The Autoresearch open-source project brings the same approach to Claude Code and Codex. Give it a goal and quantifiable metrics, …
The Autoresearch open-source project ports Karpathy's autoresearch approach to Claude Code, OpenCode, and Codex, letting AI agents automatically loop through code changes, run validation, and roll back based on goals and quantitative metrics. It provides 14 subcommands and 9 safety hooks.