Japanese AI startup, Sakana AI, has released, ‘The AI Scientist,’ which introduces the first comprehensive system for fully automatic scientific discovery, enabling foundation models such as LLMs to perform research independently.
Curated in collaboration with the Foerster Lab for AI Research at the University of Oxford, and Jeff Clune and Cong Lu at the University of British Columbia, this AI system is capable of conducting independent scientific research and communicating its findings.
The AI Scientist harnesses frontier LLMs to generate research ideas, write code, execute experiments, visualise results, and draft scientific papers. It is also dubbed as an innovative framework, representing a significant step toward fully automatic scientific discovery.
To ensure the quality of its work, the system runs a simulated review process for evaluation, mimicking the human scientific community’s peer-review process.
Remarkably, each idea generated by the AI was developed into a full paper at a cost of less than $15 per paper and to evaluate the generated papers, researchers have designed and validated an automated reviewer.
https://x.com/SakanaAILabs/status/1823178623513239992Sakana AI also recently introduced EvoSDXL-JP, an image generation model developed through Evolutionary Model Merge where it generates Japanese styles 10x faster and is available on HuggingFace with a demo for research and education.
