JevBench

Headlines · 2

1. Nov 2
   Lev reuses weights to deepen decisions through repeated computation

   Researchers from Lehigh University and other institutions proposed Lev, a decision model that uses repeated computation rather than token generation to increase reasoning depth. According to the researchers, the same weights support multiple depths, with a lightweight stopping mechanism determining how many rounds to run. The model reads option probabilities after each round, and its rounds are jointly trained. On the 231-question public JevBench, project records for the same Lev-2.6B checkpoint show accuracy rising from 63.2% after one round to 76.2% after four rounds, a 13.0 percentage-point increase without changing the checkpoint weights. The report notes that these figures are replayed project-homepage records rather than live inference results.

   DoNews · 🔥 0
2. Oct 19
   MetaDecision enters JevBench open-source leaderboard top three

   36Kr reported that MetaDecision, a structured decision model developed by the 元任务 laboratory, entered the top three of the JevBench decision-model leaderboard. As of October 17, its 31B version ranked third among 129 ranked open-source systems, and its 12B version ranked first among models of the same size. Citing GPU measurements by the leaderboard operator, the report said the 31B version's median latency per judgment was under 0.2 seconds, excluding network time. The report describes the laboratory as focused on Agent Infra and decision models, offering an embeddable agent runtime plus knowledge and model services.

   36氪快讯 · 🔥 0

Experience and discussion from the community

Share my JevBench experienceAsk about JevBench