AI 雷達 / 週報
按原始发布周整理来源摘录与阅读线索，不自动生成个人观点或趋势结论。
重點訊號
本週動態 / 來源訊號
AI 雷達 / 訊號 · 智慧代理 · 論文 ·
來源摘錄
Large language models (LLMs) increasingly generate Markdown that is consumed by renderers, agents, code extractors, and structured downstream pipelines. Yet existing evaluations…
AI 雷達 / 訊號 · 智慧代理 · 論文 ·
來源摘錄
Social-deduction games such as Werewolf are increasingly used to evaluate LLM agents, but existing evaluations often rely on final game outcomes. We propose a belief-shift evalu…
本週動態 / 來源訊號
AI 雷達 / 訊號 · 具身智慧 · 論文 ·
來源摘錄
Human video datasets have emerged as a compelling alternative to expensive real-robot data, offering rich diversity at scale. To bridge the human-to-robot embodiment gap, existi…
AI 雷達 / 訊號 · 具身智慧 · 論文 ·
來源摘錄
Diffusion-based robot policies have become widely used in robotic manipulation, where they are typically trained with behavior cloning. However, policies trained purely from dem…
值得追蹤