- formatting
- images
- links
- math
- code
- blockquotes
- external-services
•
•
•
•
•
•
-
The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
This Apple paper reveals a shocking truth: reasoning models aren't always better with more complexity! Research shows that when problems exceed a certain complexity threshold, performance actually declines even with more thinking time. This challenges our basic understanding of AI reasoning capabilities.
-
[中文版] The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
這篇Apple論文發現了一個驚人事實：推理模型並非越複雜越好！研究顯示當問題超過某個複雜度後，即使給模型更多時間思考，表現反而會下降。這挑戰了我們對AI推理能力的基本認知。
-
a post with plotly.js
this is what included plotly.js code could look like
-
a post with image galleries
this is what included image galleries could look like
-
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
Stanford's RAPTOR method builds hierarchical text structures through recursive clustering and summarization to solve semantic information loss in traditional RAG text chunking. The approach uses soft clustering, UMAP dimensionality reduction, and Bayesian Information Criterion to optimize retrieval effectiveness while preserving complete semantics.
