📰 科技脉搏
2026-09-24
🏢 公司追踪
Anthropic 官方动态：Claude discovers a novel enzyme system with CRISPR-like repeats…
📅 09-23📎 Anthropic Blog
OpenAI Academy 两周年，回顾其以免费课程、工作坊和社区项目，帮助全球开发者、教育者及组织掌握 AI 工具。它降低学习门槛，推动负责任使用，展示学员成果与未来扩展，促进 AI 普惠。
📅 09-23📎 OpenAI Blog
Sam Altman 在联合国安理会讲话，阐述 AI 对全球安全、经济与治理的深远影响，呼吁国际社会加强合作，建立包容、安全、可问责的治理框架，确保技术惠及全人类，并防范滥用与系统性风险。
📅 09-23📎 OpenAI Blog
🔥 AI & 科技
AI 周报，深度分析行业趋势：Building AI Assistants with On-Device Memory…
📎 The Batch
Artificial intelligence offers an unprecedented opportunity to augment human capabilities, yet progress at the frontier has focused primarily on advancing model capabilities. We introduce StudentBench, a suite of AI teaching evaluations and a public platform that enables large-scale data collection
📅 09-23📎 arXiv AI Papers
Social navigation typically assumes a specified goal and focuses on reaching it while respecting social conventions, whereas robot group joining requires predicting where to join based on the group’s real-time activity and formation. This is a highly semantic task, yet an important capability for ap
📅 09-23📎 arXiv AI Papers
Large language models (LLMs) are increasingly used in coding tasks, but their ability to reason about code execution remains unclear. Existing repository-level QA benchmarks mainly evaluate static code understanding and often rely on LLM-based evaluation, while execution-reasoning benchmarks are mos
📅 09-23📎 arXiv AI Papers
研究数学推理模型中的顺序效应：最终答案常对条件或选项顺序保持不变，但内部表征却高度顺序敏感。作者通过排列扰动与表示探测，揭示顺序信息在推理过程中被编码，为鲁棒性评估和提示设计提供新视角。
📅 09-23📎 arXiv AI Papers
提出面向LLM智能体的“智能体编辑式”世界模型，重新审视传统预测式建模的局限。该方法让智能体主动编辑和更新环境状态表示，而非仅预测下一状态，从而提升复杂任务规划、工具调用与长程决策的一致性和可解释性。
📅 09-23📎 arXiv AI Papers
针对潜在流世界模型中“冻结流”导致运动信息丢失的问题，诊断其遗忘机制并提出恢复方法。通过定位运动表征退化并注入或重训动态模块，恢复视频预测中的运动连贯性，提升世界模型长期动态建模能力。
📅 09-23📎 arXiv AI Papers
Vector Symbolic Algebras project data structures into a hyperdimensional vector space through the application of their vector algebras to randomly generated atomic vector symbols and fractional power encodings of real-valued data. Embedding unstructured data remains an open question. We present \tex
📅 09-23📎 arXiv AI Papers
Reinforcement learning with verifiable rewards (RLVR) supervises mathematical reasoning through final-answer correctness, but provides little guidance on individual tokens. On-policy distillation (OPD) supplies dense feedback on student-generated responses, yet teacher preference need not reflect co
📅 09-23📎 arXiv AI Papers
Consumers increasingly delegate purchasing decisions to Large Language Models (LLMs) acting as surrogate consumers. Using “Tool-Lab,” an adaptation of information-board process tracing that places product attributes behind costly tool calls, we examine how marketing pricing cues (i.e., just-below pr
📅 09-23📎 arXiv AI Papers
该研究提出 AnchorReasoning，一个面向长尾自动驾驶场景的视觉定位与因果推理数据集，旨在缓解罕见危险场景数据稀缺问题。数据集通过锚点标注关联目标定位与因果问答，支持模型理解事件成因、评估鲁棒性与可解释性，为自动驾驶长尾推理研究提供基准。
📅 09-23📎 arXiv AI Papers
💰 财经 & 投资
本文延续对 Muse 的分析，探讨其与亚马逊、沃尔玛在零售与 AI 搜索领域的竞合关系，并延伸至 Expedia 等垂直平台的潜在影响。文章进一步追问：在 Muse 式智能代理重塑流量入口的背景下，谷歌赖以生存的搜索广告模式将面临何种冲击，其护城河是否依然稳固。
📎 Stratechery
由 Hermes 自动生成 | 查看历史摘要