Research

Paper

AI LLM March 17, 2026

More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verification

Authors

Song Tae-Eun

Abstract

Cross-Context Review (CCR) improves LLM verification by separating production and review into independent sessions. A natural extension is multi-turn review: letting the reviewer ask follow-up questions, receive author responses, and review again. We call this Dynamic Cross-Context Review (D-CCR). In a controlled experiment with 30 artifacts and 150 injected errors, we tested four D-CCR variants against the single-pass CCR baseline. Single-pass CCR (F1 = 0.376) significantly outperformed all multi-turn variants, including D-CCR-2b with question-and-answer exchange (F1 = 0.303, $p < 0.001$, $d = -0.59$). Multi-turn review increased recall (+0.08) but generated 62% more false positives (8.5 vs. 5.2), collapsing precision from 0.30 to 0.20. Two mechanisms drive this degradation: (1) false positive pressure -- reviewers in later rounds fabricate findings when the artifact's real errors have been exhausted, and (2) Review Target Drift -- reviewers provided with prior Q&A exchanges shift from reviewing the artifact to critiquing the conversation itself. Independent re-review without prior context (D-CCR-2c) performed worst (F1 = 0.263), confirming that mere repetition degrades rather than helps. The degradation stems from false positive pressure in additional rounds, not from information amount -- within multi-turn conditions, more information actually helps (D-CCR-2b > D-CCR-2a). The problem is not what the reviewer sees, but that reviewing again invites noise.

Metadata

arXiv ID: 2603.16244
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-17
Fetched: 2026-03-18 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.16244v1</id>\n    <title>More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verification</title>\n    <updated>2026-03-17T08:29:29Z</updated>\n    <link href='https://arxiv.org/abs/2603.16244v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.16244v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Cross-Context Review (CCR) improves LLM verification by separating production and review into independent sessions. A natural extension is multi-turn review: letting the reviewer ask follow-up questions, receive author responses, and review again. We call this Dynamic Cross-Context Review (D-CCR). In a controlled experiment with 30 artifacts and 150 injected errors, we tested four D-CCR variants against the single-pass CCR baseline. Single-pass CCR (F1 = 0.376) significantly outperformed all multi-turn variants, including D-CCR-2b with question-and-answer exchange (F1 = 0.303, $p &lt; 0.001$, $d = -0.59$). Multi-turn review increased recall (+0.08) but generated 62% more false positives (8.5 vs. 5.2), collapsing precision from 0.30 to 0.20. Two mechanisms drive this degradation: (1) false positive pressure -- reviewers in later rounds fabricate findings when the artifact's real errors have been exhausted, and (2) Review Target Drift -- reviewers provided with prior Q&amp;A exchanges shift from reviewing the artifact to critiquing the conversation itself. Independent re-review without prior context (D-CCR-2c) performed worst (F1 = 0.263), confirming that mere repetition degrades rather than helps. The degradation stems from false positive pressure in additional rounds, not from information amount -- within multi-turn conditions, more information actually helps (D-CCR-2b &gt; D-CCR-2a). The problem is not what the reviewer sees, but that reviewing again invites noise.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <published>2026-03-17T08:29:29Z</published>\n    <arxiv:comment>10 pages, 2 figures</arxiv:comment>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Song Tae-Eun</name>\n    </author>\n  </entry>"
}