The authors assembled 50 problems contributed by specialists across theoretical physics.
Frontier AI agents solve most expert-level theoretical physics calculations
On 50 unpublished, verifiable research problems, the strongest tested agent reached 98 percent accuracy with literature search and computational tools.
Top University
A. Ramesh Chandra · Aaghaz Mahajan · Ainesh Sanyal · Mianqi Wang · Shubhadeep Chakraborty · Sudip Chakraborty · +29 more
International Centre for Theoretical Sciences · Tata Institute of Fundamental Research · University of Texas at Austin · Kavli IPMU (WPI), The University of Tokyo · Harish-Chandra Research Institute
Research Digest··2 min read
Chandra and colleagues tested autonomous AI agents on research-level calculations spanning theoretical physics, using problems whose answers were not publicly available during evaluation.
Why this paper
From Kavli IPMU (WPI), The University of Tokyo and 12 others
In one line
Frontier AI agents solve hard theoretical physics problems with 98% accuracy at low cost, matching world experts.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors (3 noted)
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§