一步错步步错？实测三类信用信号在LLM智能体训练中均无法锁定关键步骤，揭示因果归因盲区。
arXiv:2608.19760v1 Announce Type: new Abstract: Audited against causal ground truth from executed replay in a single-agent tool environment (ALFWorld)…
大模型理解因果难题？这个智能体框架专攻因果推理，让LLM真正掌握因果逻辑
arXiv:2408.06849v3 Announce Type: replace Abstract: The large language model (LLM) has achieved significant success across various domains. However, t…
用因果推断给数字孪生做“证伪测试”，让仿真模型不再自说自话。
arXiv:2301.07210v5 Announce Type: replace-cross Abstract: Digital twins are simulation-based models designed to predict how a real-world process will …
因果发现也能讲人话？GENESIS提出可解释方法，给复杂因果推断配上一份清晰说明书，值得研究因果推理的读者细品。
arXiv:2608.03868v1 Announce Type: new Abstract: Causal Discovery (CD) from observational data faces two fundamental challenges. First, purely statisti…
提出显式可审计的图推理方法，通过目标感知的因果链构建实现推理路径透明化
arXiv:2607.15281v1 Announce Type: new Abstract: Causal and intervention-based question answering is fundamental to advancing large language models (LL…
如何从非结构化数据中发现因果？本文提出利用领域知识增强因果发现的新方法，突破传统局限。
arXiv:2607.09348v1 Announce Type: new Abstract: Causal discovery from unstructured data is a challenging yet underexplored task in high-expertise doma…
人机协作下个性化因果纠偏方案，让AI决策建议更贴合个体处境，XAI领域前沿探索。
arXiv:2607.03425v1 Announce Type: new Abstract: Algorithmic recourse addresses the challenge of providing tailored recommendations to users affected b…
用游戏场景量化LLM智能体的因果推理能力，CausalGame提供全新基准测试框架。
arXiv:2607.04293v1 Announce Type: cross Abstract: Building AI Scientist agents with Large Language Models (LLMs) has recently attracted growing attent…
从结果标签升级到因果过程监督，OpenRCA 2.0带来可解释的AI推理新范式。
arXiv:2606.27154v1 Announce Type: new Abstract: Root cause analysis (RCA) poses a holistic test of LLM agentic capabilities, such as long-context unde…
当LLM被捧为决策助手时，这份研究揭露了它们“因果谨慎”会在特定语境下悄然失效，是AI安全与可信度的重要警示。
arXiv:2606.24370v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into decision-support roles in business and p…
这篇论文从机制层面探究大模型检测自杀内容的内在工作原理，并提出让因果断言更可信的方法，专业深度十足。
arXiv:2606.21078v1 Announce Type: new Abstract: Large language models are increasingly proposed for mental-health applications such as detecting suici…
突破生成模型因果盲区，ARIA以因果感知框架重塑材料发现中的大模型推理，KDD 2026前沿成果。
arXiv:2606.22375v1 Announce Type: new Abstract: Generative models have revolutionized the process of materials discovery, yet they often fail to satis…
图数据常违反独立同分布假设，这项研究给出因果视角下的公平节点分类新方案，值得关注。
arXiv:2505.01652v2 Announce Type: replace-cross Abstract: Fair machine learning seeks to identify and mitigate biases in predictions against unfavorab…
智能体时代如何发现因果关系？这篇最新论文给出前沿方法框架。
arXiv:2606.23608v1 Announce Type: new Abstract: Recent attempts to combine large language models (LLMs) with causal discovery ask models to infer pair…
新方法GIST-CMTF让LLM智能体精准推断目标状态，仅暴露因果必要工具，大幅降低选择混淆。
arXiv:2606.16813v1 Announce Type: new Abstract: Tool-augmented LLM agents rely on runtime filtering to decide which tools should be visible at each st…
别让LLM直接说话，探知其隐藏状态才能获得更深层的因果推理能力。
Article URL: https://blog.j11y.io/2026-06-10_hidden-state-probes/ Comments URL: https://news.ycombinator.com/item?id=48498283 Points: 7 # Comments: 0
用轻量级合约教会LLM代理何时该用工具以及工具产生什么效果，提升可靠性与推理能力。
arXiv:2606.07904v1 Announce Type: new Abstract: Tool-augmented large language model agents increasingly rely on external APIs, but standard tool schem…
因果迁移学习如何提升医学图像分析的鲁棒性和可解释性？这篇论文给出系统性探索。
arXiv:2603.24388v2 Announce Type: replace Abstract: Medical imaging models frequently fail when deployed across hospitals, scanners, populations, or i…
用反事实归因精准定位LLM Agent失败的根本决策步骤，超越简单启发式方法，让调试不再只看表面。
arXiv:2606.08275v1 Announce Type: new Abstract: When an LLM agent fails -- issues a refund it should not have, calls the wrong tool, leaks data -- exi…
融合因果推理与LLM，自动检测ML模型公平性并生成分析报告，直击AutoML公平性盲区
arXiv:2604.27011v2 Announce Type: replace-cross Abstract: AutoML, intended as the process of automating the application of machine learning to real-wo…