← back to paper

Review history

arxiv: 2607.01240 · 2 revisions

Prompt Framing Distorts Count Based Evaluation of LLM Error Detection: Evidence from Numeric Anchoring

1. 2026-08-02 CONDITIONAL HIGH v1.3.0-alltime-deepseek novelty 5.0
   205817 ms 15819 in 22864 out 2026-08-02T14:58:43.654222+00:00
2. 2026-07-04 UNVERDICTED LOW v0.9.1-grok novelty 7.0
   30821 ms 5742 in 1389 out 2026-07-04T01:48:40.257131+00:00