https://ai-tokens.cn/news/news-meta-muse-20260922.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Meta 个人 AI 智能体 Muse 上线两周登顶 App Store，单日市值飙升近 2000 亿美元 https://ai-tokens.cn/news/news-gpt-6-astra-20260903.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 OpenAI 发布 GPT-6 Astra：「欢迎来到 AGI 时代」，API 定价 $10/$50 为上代 2.5 倍 https://ai-tokens.cn/news/news-gemini-38-flash-20260903.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Google 三周内再推 Gemini 3.8 Flash：介绍价 $0.75/$3.75，Cyber 网络安全变体齐发 https://ai-tokens.cn/news/news-workbuddy-open-platform-20260903.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 WorkBuddy 开放平台解析：用 Skill、Connector、专家与知识库给 AI 同事「配工具」 https://ai-tokens.cn/news/news-wechatpay-ai-card-20260831.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 微信支付 AI 专属卡上新：一张卡直连 DeepSeek Harness 与 OpenClaw https://ai-tokens.cn/news/news-anthropic-mhs-20260828.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Anthropic 发布 MHS 模型硬件标准：让 AI 像插 USB-C 一样操控真实世界设备 https://ai-tokens.cn/news/news-939d943fc8f0d399.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 How OpenAI turned its coding tools into dots, an always-on personal assistant https://ai-tokens.cn/news/news-8af7b30500f9c04c.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Cohere says Alibaba’s Qwen model tried to send a file to Beijing https://ai-tokens.cn/news/news-bb157bb6b3d38a63.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Phronix: The Capability-Centric Learning Paradigm for Agents https://ai-tokens.cn/news/news-8c3de05bbbcf22cf.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Patronizing and Condescending Language Detection in English and Mandarin https://ai-tokens.cn/news/news-929adda0efb128e5.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Don't Forget What You Just Learned: Self-Directed Repair Turns Agent Mistakes Into Lasting Lessons https://ai-tokens.cn/news/news-0011f44db0acd064.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 TACS-RF: Tool-driven Agentic Circuit Synthesis of RF and mmWave Modules https://ai-tokens.cn/news/news-b4a230b7efdc5793.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Permission to Retry: A Study of Authorization and Recovery in AI Agent Workflows https://ai-tokens.cn/news/news-22f689f1c4c7ac34.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Beyond On-Policy vs. Off-Policy: At What Level Should LLM Agent Distillation Follow the Student? https://ai-tokens.cn/news/news-94ca820f73fd50fd.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 $M^3$: A Multi-LLM System for Multi-Turn Jailbreaking as a Markov Decision Process https://ai-tokens.cn/news/news-31654d812ef6de5e.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Knowing Whom to Trust is not Enough: The Trust--Agreement Gap in Repeated Multi-Party LLM Negotiation https://ai-tokens.cn/news/news-4c95a0869c900b4e.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Reproducible Evaluation as the Substrate for Safe Prompt Compression in Production Agents https://ai-tokens.cn/news/news-863d73f768e26cca.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 The Helpful Agent Problem: When Emergent Cooperation Among LLM Agents Bypasses Intended Workflows https://ai-tokens.cn/news/news-32ae6d5dafe0bc18.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 CEMIR vs. FISER: a comparative study of reasoning-based and enrichment-based approaches to pragmatic intent recognition https://ai-tokens.cn/news/news-2f368a303e9112aa.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 BARC-Bench: Benign Asks, Risky Consequences https://ai-tokens.cn/news/news-0a8f2fad6cda449e.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 RSI-Opt: Recursive Self-Improvement Optimizer Agents For Model Training https://ai-tokens.cn/news/news-d0a8eada7cef3cd1.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Validating the Comparison, Not Just the Scorer: Why Overall Accuracy Can Mislead in Interactive LLM Evaluation https://ai-tokens.cn/news/news-e8d538cc87f83269.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 MisconceptDialogue: Evaluation Criteria Shape Measured Success in LLM Math Tutoring https://ai-tokens.cn/news/news-8d2b109a7e0f631a.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Rethinking Structured Generation: Can Graph-Based Reasoning Resolve Ambiguity? https://ai-tokens.cn/news/news-ea6625878e572d72.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 The Schema Coherence Inventory: Measuring Whether an Assigned Belief Organizes a Language Model's Cognition https://ai-tokens.cn/news/news-b66fa642e082d66a.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Knowledge Graphs Quality Assessment in the Era of LLMs: A Conceptual Framework https://ai-tokens.cn/news/news-93dd69343b08741d.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 An Agentic Approach with Evidence-Grounded Verification for Multimodal Emotion Reasoning in Conversation https://ai-tokens.cn/news/news-cdf1cc73b0dad3ba.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Distributional Alignment of LLMs and Its Generalization Properties https://ai-tokens.cn/news/news-46d59ee5b121c7a4.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 LAMB: A Modern Contextual Language Model for Latin https://ai-tokens.cn/news/news-4b3023696b4daa19.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Free Lunch for LLMs Confidence: The RL Critic Already Knows https://ai-tokens.cn/news/news-b696ec5a243c2b0f.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Latent Coherence: Do Frontier Language Models' Probability Judgments Obey the Law of Total Probability? https://ai-tokens.cn/news/news-16eaf7ff12fe97d0.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 SAGE: Sense-Aligned Multimodal Grounding Evaluation with Gloss-Augmented Reasoning https://ai-tokens.cn/news/news-5be8010157aca113.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 SPECMEM: Accelerating Agentic Tool Calling via Live Memory Management https://ai-tokens.cn/news/news-c27f0a7a56ce573d.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Not Every Agent Needs the Strongest Model: Unified Prompt-Model Optimization for Heterogeneous LLM-based Multi-Agent Systems https://ai-tokens.cn/news/news-45272fb4d92a8982.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Toward an Agentic AI Approach to Public Health Data Collection https://ai-tokens.cn/news/news-6dd04995b52511f6.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Agent-Based Transport Model Calibration via Online Stochastic Batch Optimization https://ai-tokens.cn/news/news-ec7b65385c91a7fc.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Emergent Cooperative Thermoregulation in Lunar Swarm Robotics https://ai-tokens.cn/news/news-2f9fa02c320e889c.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Distributed Swarm Morphogenesis with Neural Cellular Automata https://ai-tokens.cn/news/news-5228f74672fd7e8b.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Agentic AI Defense: Decode-Time Provenance Binding for MCP https://ai-tokens.cn/news/news-87e89f01eabe8993.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 PULSES: multi-agent AI for autonomous excited-state model building and dynamics https://ai-tokens.cn/news/news-c5963ff2d48b21f9.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 HGWM: Hierarchical Graph-Guided World Model for Zero-Shot Object Navigation via Graph Matching https://ai-tokens.cn/news/news-3ab669561f9fc06b.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 J-Comm: Jacobian-Guided Latent Communication between LLM Agents https://ai-tokens.cn/news/news-9574e236fec705f5.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Do "True" and "False" Add Up to One? Complement Consistency in LLM-Agent Opinion Simulations https://ai-tokens.cn/news/news-dd66b9dcc37d8f3f.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 BeforeIT.jl: High-Performance Agent-Based Macroeconomics Made Easy https://ai-tokens.cn/news/news-c15b1e3c66d84564.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 GPT-6.1 Terra https://ai-tokens.cn/news/news-b1d7eb2dedce5839.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Our newest AI Quest: Mapping the human brain https://ai-tokens.cn/news/news-7d972df501d01ba1.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 GPT-6.1 Sol helps CodeRabbit strengthen security across complex codebases https://ai-tokens.cn/news/news-742d30efdc4090c1.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 How Ultrafast cut incident investigation from hours to minutes at OpenAI https://ai-tokens.cn/news/news-744292a2af160189.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 GPT-6.1 Terra halves OpenAI's API prices and still gains ground on its larger models https://ai-tokens.cn/news/news-118a0be49606e101.html AI导航 zh-CN 2026-10-27T11:19:06+08:00 Anthropic opens its books for Wall Street, showing record revenue and mounting obligations