Open-source LLM simulator separates structural effects from LLM quirks in benchmark markets
sanmikoyejo · x · 2026-10-11
The authors frame their benchmark-market simulator as a stylized model and hypothesis generator. A rule-only mode runs the same world without LLMs, separating structural effects from LLM-specific quirks — and the direction of the findings holds there too. Code is open, and the site lets you click any actor to inspect what it observes.
More from Research
- Toronto surgeons train AI to flag safe incision zones in real time during surgery — EricTopol · 2026-10-11
- Mathematician digests OpenAI's number theory results; Hodge papers pulled over sign error — lpachter · 2026-10-11
- SpIDER paper boosts code retrieval for coding agents via semantic search plus code graphs — mangahomanga · 2026-10-11
- CMU professor builds detailed 3D dragon from 27KB of code via Astra — 141_1337 · 2026-10-11
- Claude surfaces hidden planetary system 158 light-years away from public telescope data — DavidmComfort · 2026-10-11
- Diffusion LM Best-Paper Author Dropped Out of Stanford PhD to Join OpenAI — aaron_lou · 2026-10-11