AI.news
主页教程研究工具模型AI创业讨论新闻每日简报WIKI│🚀 创业库★ 投稿
AI+医疗机器人教育金融能源健康娱乐思考
181
0
研究The Sequence Knowledge - Issue 911: Distilling Diffusion and Multimodal Models(thesequence.substack.com)
14天前原文·0 评论·
0
评论
182
0
研究NL2SHACL-Bench: A Benchmark Suite for Natural Language to SHACL Translation(arxiv.org)
14天前原文·0 评论·
0
评论
183
0
研究Flow-by-Flow:Content-Judgment Bypass for Governing AI Output in High-Loss Domains(arxiv.org)
14天前原文·0 评论·
0
评论
184
0
研究MetaSpace: Metamorphic Testing for Spatial Cognition in Embodied Agents(arxiv.org)
14天前原文·0 评论·
0
评论
185
0
研究TREAT: Evaluating Access to Formal Knowledge across Equivalent Mathematical Representations(arxiv.org)
14天前原文·0 评论·
0
评论
186
0
研究Towards an Argumentative Foundation for Evaluative AI(arxiv.org)
14天前原文·0 评论·
0
评论
187
0
研究Dynamic Coalition Formation and Communication Pricing in Skill-Based Agentic AI Systems(arxiv.org)
14天前原文·0 评论·
0
评论
188
0
研究The Knowing-Saying Gap: When Probes See Errors that Confidence Misses(arxiv.org)
14天前原文·0 评论·
0
评论
189
0
研究When LLM Agents Negotiate: Private Information and Dynamic Bargaining in Supply Chains(arxiv.org)
14天前原文·0 评论·
0
评论
190
0
研究Training Variable Long Sequences with Data-Centric Parallel(arxiv.org)
14天前原文·0 评论·
0
评论
191
0
研究AI professors are negotiating the new realities of academic research(technologyreview.com)
14天前原文·0 评论·
0
评论
192
0
研究They said they would build AI safely. Then it went rogue.(cset.georgetown.edu)
14天前原文·0 评论·
0
评论
193
0
研究5 useful things you'll learn in my new post-training textbook (shipping now!)(interconnects.ai)
15天前原文·0 评论·
0
评论
194
0
jack-clark.net
研究Import AI 468: 23 RSI ideas; PostTrainBench+; and how trust and transparency interplay with AI racing(jack-clark.net)
15天前原文·0 评论·
0
评论
195
0
研究The Download: AI agents for science, and the “censorship-industrial complex”(technologyreview.com)
15天前原文·0 评论·
0
评论
196
0
研究AI for science needs reasoning, not just data(technologyreview.com)
15天前原文·0 评论·
0
评论
197
0
研究Divergent Response Modes in Frontier Language Models Under Steering Pressure(arxiv.org)
15天前原文·0 评论·
0
评论
198
0
研究Beyond Routing Weights: Faithful Response-Level Interpretation of Mixture-of-Experts Reward Models via Contribution Contrast(arxiv.org)
15天前原文·0 评论·
0
评论
199
0
研究Can MLLMs Decode the Creative Leap? Introducing C4 for Cross-Concept Understanding(arxiv.org)
15天前原文·0 评论·
0
评论
200
0
研究WebGrader: Training LLMs for Web Development with Self-Evolving Programmatic Grader(arxiv.org)
15天前原文·0 评论·
0
评论
上一页1234567