弱模型协作时，共享记忆会放大幻觉，导致推理失败。
Diagnosing Failure Modes of Shared-State Collaboration in Resource-Constrained Visual Agents
- 通过读-写-验证循环追踪视觉问答中的信息流
- 4B-8B模型在多页文档中幻觉率上升37%
- 适合研究低资源智能体协作可靠性的开发者
模块化视觉推理系统越来越多依赖共享工作内存进行多步协作，但低容量场景下中间状态演化的失败机制仍不明确。本文通过噪声累积视角研究弱模型（4B–8B参数量）协作推理的失败模式。提出CoSee审计框架，形式化读-写-验证循环以追踪文档视觉问答中的信息流。在跨多页文档、图表和网页的基准测试中，发现反直觉现象：简单共享空间常加剧幻觉而非缓解。识别出两种主导失败模式：噪声强化（无根据笔记被重复引用）与策略坍缩（上下文增加导致答案趋短且不明确）。基于成本-精度帕累托前沿分析表明，缺乏显式验证时，计算资源增加可能反而降低性能。结果表明，对资源受限代理而言，瓶颈不在推理深度，而在通信保真度，为可靠模块化设计提供了可追溯的诊断工具与机制基线。
原文摘要 · Abstract (English)
Modular visual reasoning systems increasingly rely on shared working memory for multi-step collaboration, yet the failure dynamics of intermediate state evolution in low-capacity regimes remain underexplored. We study failure modes of collaborative reasoning with weak learners (4B--8B models) through the lens of noise accumulation. We introduce CoSee, an auditing framework that formalizes the read-write-verify loop to trace information flow in document visual question answering. Across multi-page, chart, and web-based benchmarks, we find a counter-intuitive degradation: naive shared workspaces often amplify hallucinations rather than resolve them. We identify two dominant failure modes: Noise Reinforcement, where ungrounded notes are reused as evidence, and Policy Collapse, where added context shifts the model toward under-specified, short-form answers. Using cost-accuracy Pareto frontiers, we show that increased compute can correlate negatively with performance without explicit verification. Our findings suggest that for resource-constrained agents, the bottleneck lies not in reasoning depth but in communication fidelity, providing trace-level diagnostics and a mechanistic baseline for reliable modular design.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。