, cloth, fluid, rigid body).
LLM agents struggle to generate physics simulators from descriptions, new benchmark shows
WorldSolver benchmark of 168 tasks from computer graphics papers tests agents on execution, visual fidelity, and physical plausibility of generated solvers.
Chinese Tech
Siru Jiang · Yongzhe Lyu · Shuo Lu · Yubin Wang · Yuxiang Zhang · Yue Liao · +3 more
Chinese Academy of Sciences · Peking University · Huawei
Research Digest··3 min read
The authors introduce WorldSolver, a benchmark of 168 simulation tasks derived from physical phenomena in 61 classic computer graphics papers across 7 domains.
Why this paper
From Chinese Academy of Sciences and 2 others
In one line
Generating correct physics solvers from text descriptions is difficult for LLM agents, as shown by the WorldSolver benchmark.
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
- ✓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.
§