发布日期：2026-04-11
收录条目：20
1. 20-year-old man arrested for allegedly throwing a Molotov cocktail at Sam Altman’s house
- 来源：The Verge AI
- 发布时间：2026-04-10 20:16 UTC
- 链接：https://www.theverge.com/ai-artificial-intelligence/910393/openai-sam-altman-house-molotov-cocktail
摘要：San Francisco police have arrested a 20-year-old man suspected of throwing a Molotov cocktail at OpenAI CEO Sam Altman's Russian Hill house early Friday morning, The San Francisco Standard reports. The incident was caugh
2. The Iranian Lego AI video creators credit their virality to ‘heart’
- 来源：The Verge AI
- 发布时间：2026-04-10 17:30 UTC
- 链接：https://www.theverge.com/ai-artificial-intelligence/909948/explosive-media-lego-iran-war-trump-netanyahu
摘要：Donald Trump has spun the recent rescue of a downed airman whose fighter jet was destroyed behind Iranian borders as a resounding success. But the story is very different in one of the many viral, AI-generated Lego video
3. Fear and loathing at OpenAI
- 来源：The Verge AI
- 发布时间：2026-04-10 12:23 UTC
- 链接：https://www.theverge.com/podcast/909621/openai-sam-altman-drama-vergecast
摘要：Sam Altman's tenure at OpenAI has been… messy. Messy to the point where Altman was briefly fired from his role as CEO, only to be reinstated days later, at which point he began reshaping the organization permanently. Thi
4. Gen Z’s love-hate relationship with AI
- 来源：The Verge AI
- 发布时间：2026-04-10 11:23 UTC
- 链接：https://www.theverge.com/ai-artificial-intelligence/909687/gen-z-doesnt-like-ai-gallup
摘要：Gen Z is increasingly disillusioned with AI - just not enough to stop using it. A new Gallup report released this week, based on responses from nearly 1,600 people ages 14 to 29 across the US, suggests the hype is wearin
5. Microsoft starts removing Copilot buttons from Windows 11 apps
- 来源：The Verge AI
- 发布时间：2026-04-10 09:22 UTC
- 链接：https://www.theverge.com/news/909640/microsoft-removing-copilot-windows-11-buttons
摘要：Microsoft is starting to remove "unnecessary" Copilot buttons from its Windows 11 apps. In the latest version of the Notepad app for Windows Insiders, Microsoft has removed the Copilot button in favor of a "writing tools
6. High-Precision Estimation of the State-Space Complexity of Shogi via the Monte Carlo Method
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06189
摘要：arXiv:2604.06189v1 Announce Type: new Abstract: Determining the state-space complexity of the game of Shogi (Japanese Chess) has been a challenging problem, with previous combinatorial estimates leaving a gap of five ord
7. Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06233
摘要：arXiv:2604.06233v1 Announce Type: new Abstract: Safety-trained language models routinely refuse requests for help circumventing rules. But not all rules deserve compliance. When users ask for help evading rules imposed b
8. Toward Reducing Unproductive Container Moves: Predicting Service Requirements and Dwell Times
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06251
摘要：arXiv:2604.06251v1 Announce Type: new Abstract: This article presents the results of a data science study conducted at a container terminal, aimed at reducing unproductive container moves through the prediction of servic
9. Weakly Supervised Distillation of Hallucination Signals into Transformer Representations
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06277
摘要：arXiv:2604.06277v1 Announce Type: new Abstract: Existing hallucination detection methods for large language models (LLMs) rely on external verification at inference time, requiring gold answers, retrieval systems, or aux
10. SymptomWise: A Deterministic Reasoning Layer for Reliable and Efficient AI Systems
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06375
摘要：arXiv:2604.06375v1 Announce Type: new Abstract: AI-driven symptom analysis systems face persistent challenges in reliability, interpretability, and hallucination. End-to-end generative approaches often lack traceability
11. SELFDOUBT: Uncertainty Quantification for Reasoning LLMs via the Hedge-to-Verify Ratio
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06389
摘要：arXiv:2604.06389v1 Announce Type: new Abstract: Uncertainty estimation for reasoning language models remains difficult to deploy in practice: sampling-based methods are computationally expensive, while common single-pass
12. Qualixar OS: A Universal Operating System for AI Agent Orchestration
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06392
摘要：arXiv:2604.06392v1 Announce Type: new Abstract: We present Qualixar OS, the first application-layer operating system for universal AI agent orchestration. Unlike kernel-level approaches (AIOS) or single-framework tools (
13. ProofSketcher: Hybrid LLM + Lightweight Proof Checker for Reliable Math/Logic Reasoning
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06401
摘要：arXiv:2604.06401v1 Announce Type: new Abstract: The large language models (LLMs) might produce a persuasive argument within mathematical and logical fields, although such argument often includes some minor missteps, incl
14. BDI-Kit Demo: A Toolkit for Programmable and Conversational Data Harmonization
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06405
摘要：arXiv:2604.06405v1 Announce Type: new Abstract: Data harmonization remains a major bottleneck for integrative analysis due to heterogeneity in schemas, value representations, and domain-specific conventions. BDI-Kit prov
15. On Emotion-Sensitive Decision Making of Small Language Model Agents
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06562
摘要：arXiv:2604.06562v1 Announce Type: new Abstract: Small language models (SLM) are increasingly used as interactive decision-making agents, yet most decision-oriented evaluations ignore emotion as a causal factor influencin
16. Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06628
摘要：arXiv:2604.06628v1 Announce Type: new Abstract: A prevailing narrative in LLM post-training holds that supervised finetuning (SFT) memorizes while reinforcement learning (RL) generalizes. We revisit this claim for reason
17. KD-MARL: Resource-Aware Knowledge Distillation in Multi-Agent Reinforcement Learning
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06691
摘要：arXiv:2604.06691v1 Announce Type: new Abstract: Real world deployment of multi agent reinforcement learning MARL systems is fundamentally constrained by limited compute memory and inference time. While expert policies ac
18. Reasoning Fails Where Step Flow Breaks
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06695
摘要：arXiv:2604.06695v1 Announce Type: new Abstract: Large reasoning models (LRMs) that generate long chains of thought now perform well on multi-step math, science, and coding tasks. However, their behavior is still unstable
19. AgentGate: A Lightweight Structured Routing Engine for the Internet of Agents
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06696
摘要：arXiv:2604.06696v1 Announce Type: new Abstract: The rapid development of AI agent systems is leading to an emerging Internet of Agents, where specialized agents operate across local devices, edge nodes, private services,
20. ATANT: An Evaluation Framework for AI Continuity
- 来源：arXiv cs.AI
- 发布时间：2026-04-10 04:00 UTC
- 链接：https://arxiv.org/abs/2604.06710
摘要：arXiv:2604.06710v1 Announce Type: new Abstract: We present ATANT (Automated Test for Acceptance of Narrative Truth), an open evaluation framework for measuring continuity in AI systems: the ability to persist, update, di