Re-Search: A 20-Questions Benchmark for LLM Research Ideation
Researchers from Texas A&M and JHU introduced Re-Search, a benchmark adapting the twenty-questions game to study research ideation in LLMs, finding models can recognize good ideas but struggle to generate them.
2026-11-17 ~ 2026-11-17 · 2 related posts
- Re-Search paper: LLMs can judge good research ideas but struggle to generate them — DanielKhashabi · 2026-11-17
- Re-Search benchmark shows LLMs can spot good research ideas but can't generate them — Muhan Gao · 2026-11-17