Pith. sign in

← back to paper

Review history

arxiv: 2605.13841 · 2 revisions

EVA-Bench: A New End-to-end Framework for Evaluating Voice Agents

  1. 2026-06-30 UNVERDICTED LOW v0.9.1-grok novelty 7.0
    32784 ms 5902 in 1574 out 2026-06-30T21:16:17.113275+00:00
  2. 2026-05-14 ACCEPT MODERATE v0.9.0 novelty 8.0
    71341 ms 5667 in 1513 out 2026-05-14T17:26:52.036391+00:00