AI资讯日报 - 2026/10/27

👨‍🔬 Benjamin Pikus, Daniel Saunders, Mircea Mironenco, Salman Mohammadi, Mathew Mathew, Sirou Zhu, Divya Makkar, Xuhua Huang, Qinbo Li, Supreeth Narasimhaswamy, Siyuan Cen, Thomas Neff, Dejia Xu, Shyamal Buch

Patient-specific rubric-based evaluation of large language models across longitudinal cancer care

学术论文 ArXiv 重要度: 8

NPC-RubricBench基于405名患者纵向记录评估30个LLM，发现诊断强不代表下游护理强，失败模式随治疗阶段反转。

👨‍🔬 Wei-Wei Zhang, Pengkai Wang, Yan Li, Zhijie Sang, Guan-Qun Zhou, Jun-Yan Li, Shu-Han Zhou, Pei-Wen Liu, Ting-Xi Tang, Min Tang, Zhitian Hou, Jianmin Wu, Jian Guan, i Lin, Hongxia Yang, Ying Sun

Free to Look, Free to Reason: Simple LLM Harnesses Outdiagnose Hand-Crafted Root Cause Analysis Agents

学术论文 ArXiv 重要度: 8

在RCAtlas上，简单LLM harness的根因分析准确率65%至77%，显著优于手工RCA智能体和传统统计因果方法。

👨‍🔬 Tingzhu Bi, Xun Zhang, Jinglin Li, Ping Wang, Meng Ma

Profiling a Frontier Agent with EBench: From Direct Control to Harness with Policy

学术论文 ArXiv 重要度: 8

EBench评估GPT-6-Astra控制机器人：直接控制在短时任务领先，结合后训练VLA策略可提升成功率并降低令牌成本。

👨‍🔬 Yifei Yao, Yude Zou, Guanbo Jiao, Xing Gao, Ning Gao, Yukai Wang, Jiantong Chen, Zihou Zhu, Xinyu Li, Ningyu Yan, Shuai Wang, Jianming Ma, Haoyu Guo, Weinan Zhang, Chunhua Shen

Learning Force-Reactive Humanoid Loco-Manipulation

学术论文 ArXiv 重要度: 8

FLORE用双力传感手柄采集无机器人演示，力反应扩散策略在接触密集任务中成功率80%，远超仅位置策略的26.7%。

👨‍🔬 Sihang Li, Xiaofeng Guo, Jinkun Cao, Charlie Cheng, Ruihan Yang, Chen Feng, Guanya Shi, Haozhi Qi

Learning When to Commit: Schedule Distillation for Fast Diffusion Language Models

学术论文 ArXiv 重要度: 7

dSpark通过调度蒸馏学习何时提交令牌，在GSM8K上比LLaDA-8B和Dream-7B分别加速10.12倍和5.12倍且保持性能。

👨‍🔬 Aidar Myrzakhan, Zhiqiang Shen

A Unified Arena for LLM Agents across Diverse Eurogame

学术论文 ArXiv 重要度: 6

DECE统一十款欧式桌游评估12种模型与执行框架组合，共1032场比赛，揭示跨游戏排名一致性与框架带来的性能差异。

👨‍🔬 Fengrui Zhang, Xiang Li, Ruomeng Ding, Jiayi Yuan, Yuchen Huang, Zhaoyang Han, Xiaoliang Wang, Daoyuan Chen, Yaliang Li

Game theory of academic effort and hiring in the age of AI

学术论文 ArXiv 重要度: 5

用博弈论建模AI时代学者投入深度理解与堆论文量的两难，分类纯策略均衡并模拟招聘双方可能习得的行为。

👨‍🔬 Saptarshi Pal, Peter S. Park

Center-of-Mass Interpretation of Evolution Strategies

学术论文 ArXiv 重要度: 4

给出进化策略更新的几何解释：模型向奖励加权质心移动，均值减法改变位移测量起点，对任意有限种群精确成立。

👨‍🔬 Akshat Gupta

🤖 AI资讯日报

📊 今日趋势总结

MIT Non-AI License

Ask HN: Anyone concerned about NYC Local Law 144?

Why Boring Businesses Outlast AI Hype Cycles

Ask HN: What's the pain using current AI algorithms?

The AI Crackpot Index

Ask HN: Is the rate of progress in AI exponential?

Show HN: Startup Raising capital through Book Sales

NLP, AI, ML, bots – a passing trend or much more? What's your take on this?

Ask HN: What would you read to learn about "artificial intelligence"?

Common Lisp + Machine Learning Internship at Google (Mountain View, CA)

The Next Bill Gates or Albert Einstein in AI “Chris Clark” – Yourobot

Bioinformatician

Retriever-Free Retrieval: Navigating Documents with Decision Models

I-HITL: Inalienable Human-in-the-Loop

Reva: Real-Time Vulnerability Detection through Contextual Probing of Large Language Models

Pretraining Knows What, Simulation Teaches How: Compositional Generalization in a Sim-Trained Robot Agent

Patient-specific rubric-based evaluation of large language models across longitudinal cancer care

Free to Look, Free to Reason: Simple LLM Harnesses Outdiagnose Hand-Crafted Root Cause Analysis Agents

Profiling a Frontier Agent with EBench: From Direct Control to Harness with Policy

Learning Force-Reactive Humanoid Loco-Manipulation

Learning When to Commit: Schedule Distillation for Fast Diffusion Language Models

A Unified Arena for LLM Agents across Diverse Eurogame

Game theory of academic effort and hiring in the age of AI

Center-of-Mass Interpretation of Evolution Strategies

📅 历史日报目录