全部 AI 动态

DraMaze：一种具备未来感知记忆的多智能体流水线，用于多样化的微剧剧本生成

arXiv：2610.32002v1 发布类型：新 摘要：微剧的迅猛发展使得对高质量剧本创作的需求日益增长。利用大语言模型进行长篇微剧剧本生成，既需要多样化的剧情推进，又要求具备长时段的叙事连贯性。我们的观察表明，随着生成过程的推进，候选续写内容越来越倾向于聚焦于少数几种叙事机制；而传统的检索方式则可能加剧套路的重复使用，并忽视对未来情节至关重要的历史信息。为应对这些挑战，我们提出了DraMaze，这是一种多智能体流水线，通过基于套路记忆的进化树搜索与变异机制，探索多样化的因果路径；同时，它还构建了逆向叙事依赖树，用于组织、保存并检索历史记忆。

arXiv新论文提出可追溯多模态情绪推理的智能体框架，支持证据验证与免权重更新的提示修补

2026年10月26日，arXiv发布预印本论文（编号2610.31685v1），提出一种基于冻结大模型的可追溯多模态情绪推理智能体架构，通过模态专用感知智能体、集成智能体、MLLM批评者及证据验证阶段实现错误归因与提示空间自适应优化，在MER2025数据集上提升多个开源多模态大模型（2B–38B参数）的情绪识别F-score达0.6–2.2分。

推荐理由：该研究针对当前多模态情绪推理系统在指令微调成本高与智能体流水线不可调试之间的矛盾，提出首个融合证据引用、确定性验证、失败归因与靶向提示修补的全流程可解释框架；所有改进均不修改模型权重，且在四个不同规模开源MLLM上实证有效，具备方法论创新性与工程实用性。

Permission to Retry: A Study of Authorization and Recovery in AI Agent Workflows

arXiv:2610.31916v1 Announce Type: new Abstract: A durable agent can resume a consequential action while its authority changes and the action's external outcome remains uncertain. We study execution-time reliability of tool-using AI agents, isolating recovery decisions from language-model behavior. Checkpoint state, permission to dispatch, evidence of an external effect and permission to reconcile are distinct obligations. We present a finite conformance study that checks these obligations through histories scored against a provider effect ledger hidden from the recovering application. A prospectively frozen campaign executes 336 histories on native LangGraph and Temporal hosts, crossing 14 schedules, three provider contracts and four matched application policies. All 168 paired host cells

Understanding Communication Failures in Multi-Agent LLM Systems

arXiv:2610.32060v1 Announce Type: new Abstract: Multi-agent LLM systems aim to improve reasoning through debate, evidence sharing, and mutual correction, but interaction can also introduce failures that final accuracy alone cannot explain. Agents may converge too early, ignore dissent, suppress correct minority answers, or preserve a wrong framing across rounds. We present a corpus-grounded taxonomy of communication failures, built from 3,211 failed multi-round trajectories across medical QA, coding, finance, multi-hop reasoning, ethics, and embodied text tasks. Using this taxonomy, we analyze failure distributions across datasets and system designs, round-level error dynamics, recovery hardness, majority transitions, minority-signal suppression, and self-reported confidence. We find that