跳到正文
arXiv cs.MA· Ziaullah Khan, Abdul Rehman Khalid, Shrestha Nischal Lal, Nheng Vanchhay, Md Ariful Islam Mozumder, Hee-Cheol Kim·· 9 小时前AI 评分34

DebateGraph 门控消融研究:保留分歧比消解分歧更能提升多智能体辩论表现

Surfacing Disagreement Beats Resolving It: A Gated Ablation Study of Provenance-Weighted Memory for Multi-Agent Debate

AI 导读

多智能体辩论共享记忆的 DebateGraph 门控消融研究显示,在 39 次运行中仅"保留矛盾并将排序后的冲突集传入辩论"这一机制有效:F1 提升 +0.340(95% CI [+0.267, +0.414]),122 个投毒条目中攻击成功数从 22 降至 7(McNemar 精确检验 p=0.0003),成本不变。

正文 · 原文

View PDF HTML (experimental)

Abstract:Shared memory for multi-agent debate is usually built to resolve inconsistency: retrieval returns one winning fact and the disagreement disappears before any agent sees it. We ask whether that is correct. In DebateGraph, four backends share one contract (read/write/snapshot), so five mechanisms from the agent-memory literature can be removed independently under a fixed budget, each gated on a pre-declared threshold. Across 39 runs one survives: preserving contradictions and passing the ranked conflict set to the debate, rather than collapsing it to a winner, improves F1 by +0.340 (95% CI [+0.267, +0.414]) and cuts attack success from 22 to 7 of 122 poisoned items (McNemar exact p=0.0003), at equal cost. Reaching it meant fixing a bottleneck in our harness: conflict sets are built only over retrieved edges, and our pre-declared cap left three quarters of planted contradictions unretrievable, first measuring the mechanism at +0.084. The effect replicates on two retrieval substrates and directionally on HotpotQA documents. No other mechanism survives, and two are worse than inert: once contradictions reach the prompt, a write trust-gate and the router's provenance features each cost accuracy (-0.075 and -0.096 F1). The gate fails structurally: poisoned objects are always entities the graph already contains, against 53% of gold, so groundedness scores the lie above a truthful first mention. We release the harness, attack set, and logs, and argue that silently reconciling memory is the assumption worth abandoning.
Subjects: Multiagent Systems (cs.MA); Computation and Language (cs.CL)
Cite as: arXiv:2611.22461 [cs.MA]
  (or arXiv:2611.22461v1 [cs.MA] for this version)
  https://doi.org/10.48550/arXiv.2611.22461

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Ziaullah Khan [view email]
[v1] Tue, 17 Nov 2026 09:23:13 UTC (2,941 KB)

来源:arXiv cs.MA · arxiv.org