Stanford Releases ScholarCatalyst, a Retrieval Benchmark Built From Researchers' Judgments
StanfordAILab · x · 2026-11-06
Stanford AI Lab introduces ScholarCatalyst, a reasoning-intensive retrieval benchmark for scientific discovery built from human researchers' judgments. The authors reframe retrieval as 'research intuition' and explore how it can surface promising directions for open-ended discovery.
More from Research
- Diffusion-Qwen3.8-27B: Qwen3 adapted into a masked diffusion LM with bidirectional attention — GabGarrett · 2026-11-06
- Researcher slams Inkling 2 Small time-series demo as a practically pointless task — mariofilhoml · 2026-11-06
- Wavefunction Bench Tests Coding Agents on Quantum Chemistry: Only 3 of 9 Systems Passed — CatAstro_Piyush · 2026-11-06
- Stable reinforce handles 8-minute rollouts with FP4 weights in open RL training — simonguozirui · 2026-11-06
- Biggest open RL run yet: 2,048 GPUs, 9 days, 17M rollouts, 3T+ tokens — LiTianleli · 2026-11-06
- Solv3R reconstructs 128-photo scenes with under 1.5GB of VRAM — taiyasaki · 2026-11-06