← back to paper

Review history

arxiv: 2607.06008 · 2 revisions

PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents

1. 2026-07-11 CONDITIONAL HIGH v1.1.0-grok45 novelty 6.0
   46803 ms 20430 in 3684 out 2026-07-11T01:36:03.480897+00:00
2. 2026-07-08 UNVERDICTED LOW v0.9.1-grok novelty 6.0
   48878 ms 6361 in 2838 out 2026-07-08T20:00:57.263908+00:00