When Helping Hurts and How to Fix It: Multi-Agent Debate for Data Cleaning

Computer Science > Artificial Intelligence [Submitted on 1 Jun 2026] When Helping Hurts and How to Fix It: Multi-Agent Debate for Data Cleaning Chirag Parmar, Akshat Mehta, Henglin Wu, Jagadish Ramamurthy, Shweta Medhekar When does multi-agent debate help data cleaning, and when does it hurt? Across three benchmarks, four model families, and over 6,000 task-condition pairs, we find debate's effect reverses sign: it degrades generation across all four models (-1.6 to -15.5pp) through critique-induced confusion (CIC), hallucinated Critic feedback that the Generator accepts uncritically, yet improves error detection (+27.4pp F1, d=1.0). We derive a debate benefit condition: debate helps when the probability of rescuing a wrong output (Critic verification odds weighted by fixability) exceeds the probability of destroying a correct one. A factorial experiment proves adversarial separation is essential: self-verification with identical tools fails, while a separate Critic with code-execution grounding and evidence-gated generation produces the first debate configuration to significantly exceed single-agent on a generative task (+5.3pp, p<0.05). The condition correctly predicts all nine task types and generalizes with zero false positives across 19 published comparisons in seven domains. Comments: 27 pages, 4 figures, 12 tables. Includes appendix with full experimental results, prompt templates, and dataset statistics Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA) Cite as: arXiv:2606.02866 [cs.AI]   (or arXiv:2606.02866v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2606.02866 Focus to learn more Submission history From: Chirag Parmar [view email] [v1] Mon, 1 Jun 2026 20:29:47 UTC (135 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-06 Change to browse by: cs cs.CL cs.MA References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)