同一问题换个问法，大模型对事实与信念的判断就不同，这份研究揭示了提示词如何影响模型的认知边界。
arXiv:2608.17809v1 Announce Type: new Abstract: Humans naturally form and express beliefs in daily communication, e.g., "I think the answer is 3" or "…
用信念与策略声明式控制LLM流水线，打造可解释、可调控的AI流程新范式。
arXiv:2604.14401v2 Announce Type: replace Abstract: Agentic AI systems are becoming commonplace in domains that require long-lived, stateful decision-…
专注AI研究解读的资讯站，用前沿论文帮你理解AI如何影响认知与信念。
Article URL: https://aiunderstanding.org/news/ai-chats-emerging-conspiracy-beliefs-study Comments URL: https://news.ycombinator.com/item?id=49241569 P…
英伟达低调押注AI安全，开放权重与透明审查成新战场，看懂巨头下一步棋。
IT之家 8 月 6 日消息，据《商业内幕》今天（6 日）报道，英伟达正在低调组建一支新的 AI 安全与网络安全工程团队 。 上月底集中发布的一批招聘信息首次披露了团队的存在。英伟达计划 招聘一名杰出工程师 ，出任新团队的创始技术负责人，其他岗位包括 安全研究工程师、评估工程师和高级经理 。团队将对…
六款大模型模拟人类信念更新，一对一对照实验揭示仿真失真边界，社科实验需警惕AI替身风险。
arXiv:2607.28347v1 Announce Type: new Abstract: LLMs are increasingly deployed as proxies for human study participants in social science experiments, …
提出动态认知划分下的可解释信念协调方法，融合共识迭代与逻辑推理，解决多智能体信念组合难题
arXiv:2607.21210v1 Announce Type: cross Abstract: Existing approaches to multi-agent belief combination have established mature foundations for combin…
Prompt鲁棒性并非一概而论，客观与信念式提问揭示LLM评估的新维度。
arXiv:2607.05554v1 Announce Type: cross Abstract: Survey-style evaluations of large language models often treat a prompted response as a measure of a …
多轮对话中模型信念如何随证据更新？新基准BayesBench专测LLM贝叶斯推理轨迹
arXiv:2606.30850v1 Announce Type: new Abstract: Large language models (LLMs) are typically deployed in multi-turn conversations, where each turn provi…
研究发现LLM给出的解释并不反映其真实推理过程，提出基于模型自身输入信念评估解释充分性的新方法。
arXiv:2606.28615v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in high-stakes domains, where free-text explana…
多智能体LLM的验证延迟如何破坏信念稳定？这篇论文给出稳定性阈值与最优验证器放置方案，值得算法工程师细读。
arXiv:2606.27409v1 Announce Type: cross Abstract: Multi-agent large language model (LLM) systems often rely on verifier and critic agents to suppress …
用一个SQLite文件实现AI代理的双时间记忆，支持按时间点查询信念与修正过程。
Article URL: https://github.com/las7/memharness Comments URL: https://news.ycombinator.com/item?id=48581267 Points: 1 # Comments: 1
大模型信念稳定性研究：从漂移到一致性的新方法
arXiv:2606.17832v1 Announce Type: new Abstract: Large language models (LLMs) are often hypothesized to perform implicit Bayesian inference, yet a key …
用LLM模拟社会信念演化，探索AI如何理解并预测群体认知变迁。
arXiv:2606.11482v1 Announce Type: cross Abstract: Understanding and predicting how social beliefs evolve in response to events -- from policy changes …
arXiv上的这篇论文深入探讨信念获取的随机滤波模型，免费获取前沿学术知识。
arXiv:2206.02178v3 Announce Type: replace Abstract: This paper studies how belief acquisition can be accomplished using stochastic filtering. First, a…
揭示LLM决策背后的真相：它们真的在推理还是仅仅模仿理由？这篇新研究深入探讨AI的潜意识。
arXiv:2606.11016v1 Announce Type: new Abstract: We ask whether large language models (LLMs) merely imitate rationales when choosing between two option…
融合可信推理与信念空间，用可验证神经安全过滤器实现交互式机器人的宽松安全保证。
arXiv:2606.02562v1 Announce Type: cross Abstract: Autonomous robots that interact with people must make safe and efficient decisions under human-induc…
揭秘推理模型“表演性思维链”：模型已确信答案却不暴露内部信念，引发对AI推理真实性的深层思考。
arXiv:2603.05488v4 Announce Type: replace-cross Abstract: We provide evidence of performative chain-of-thought (CoT) in reasoning models, where a mode…
研究揭示LLM在训练数据中植入错误信念后，即使明确警告也无法纠正，警示AI安全与事实性漏洞。
Fine-tuning tests show "bias ... toward confidently representing the claims as true."
研究揭示大语言模型并非始终符合贝叶斯推理，量化了其概率信念内部的不一致性。
arXiv:2605.06915v2 Announce Type: replace Abstract: Modern AI systems are being deployed in complex domains such as medicine, science, and law, where …
论文提出一种信念增强的对话生成策略，让模型在对话中主动选择澄清、拒绝或直接回答，提升智能交互的可靠性。
arXiv:2605.25831v1 Announce Type: cross Abstract: Large language models (LLMs) define a distribution over text, which can be viewed as a probabilistic…