← back to paper

Review history

arxiv: 2608.09538 · 2 revisions

TCS-BENCH: Benchmarking State-of-the-Art Generative AI Theoretical Computer Science Research Ability

1. 2026-08-14 CONDITIONAL HIGH v1.4.0-alltime-deepseek-medium novelty 7.0
   111230 ms 20450 in 12771 out 2026-08-14T04:13:57.714125+00:00
2. 2026-08-11 REJECT MODERATE v1.4.0-daily-deepseek-medium novelty 6.0
   128988 ms 11617 in 16446 out 2026-08-11T15:27:16.201423+00:00