Cao Zihang - Blog

「抵近知识无人区」

【阅】本周阅读摘选2026-06-15 → 2026-06-21

本周阅读摘选 2026-06-15 → 2026-06-21 目录 学术相关 Large language models and causal inference in collaboration: a survey 学术相关 Large language models and causal inference in collaboration: a survey 1 Treatment Effet Estimation ...

【阅】本周阅读摘选2026-06-08 → 2026-06-14

本周阅读摘选 2026-06-08 → 2026-06-14 目录 学术相关 数据派THU丨别把 LLM 提取的主题当成真实变量 技术技巧 DeepHub丨Pydantic V2 CheatSheet 学术相关 数据派THU丨别把 LLM 提取的主题当成真实变量 LLM 生成的主题标签是“生成变量”,它继承了文本产生机制、模型提取过程和样本选择机制中的隐含假设。因而,在将此类变量用于回归、A/B 测试或因果推...

【阅】本周阅读摘选2026-06-01 → 2026-06-07

本周阅读摘选 2026-06-01 → 2026-06-07 目录 学术相关 SkillAdaptor: Self-adapting skills for LLM agents from trajectories Personalized Deep Research: A User-Centric Framework, Dataset, and Hybrid Evaluation for Knowledge Discovery 学术相关 S...

【阅】本周阅读摘选2026-05-25 → 2026-05-31

本周阅读摘选 2026-05-25 → 2026-05-31 目录 学术相关 Intern-Atlas 0. Execution Overview 1. High-level Design (Indexing → Retrieval → Generation) 2. Offline Construction: Indexing (Detailed Execution) 3. Onli...

【阅】本周阅读摘选2026-05-18 → 2026-05-24

本周阅读摘选 2026-05-18 → 2026-05-24 目录 学术相关 HippoRAG 0. Execution Overview 1. High-level Design (Indexing → Retrieval → Generation) 2. Offline Construction: Indexing (Detailed Execution) 3. Online Q...

【阅】本周阅读摘选2026-05-11 → 2026-05-17

本周阅读摘选 2026-05-11 → 2026-05-17 目录 学术相关 LightRAG 0. Execution Overview 1. High-level Design (Indexing → Retrieval → Generation) 2. Offline Construction: Indexing (Detailed Execution) 3. Online Q...

【阅】本周阅读摘选2026-05-11 → 2026-05-17

本周阅读摘选 2026-05-11 → 2026-05-17 目录 学术相关 PaSa 0. Execution Overview 1. High-level Design (Indexing → Retrieval → Generation) 2. Offline Construction: Training (Detailed Execution) 3. Online Query...

【阅】本周阅读摘选2026-04-27 → 2026-05-03

本周阅读摘选 2026-04-27 → 2026-05-03 目录 学术相关 1. A Brief Review of Traditional Information Retrieval 1.1 The Evolution of Retrieval Methods 1.2 The Evolution of Ranking Methods 1.3 Pre-trained Transformers and ...

【阅】本周阅读摘选2026-04-20 → 2026-04-26

本周阅读摘选 2026-04-20 → 2026-04-26 目录 学术相关 Is AI an algorithm by any other name? Behavioral reactions to AI- and model-based demand planning algorithms Unraveling multifaceted user preferences on digital platforms: a bayesian deep-learnin...

【阅】本周阅读摘选2026-04-13 → 2026-04-19

本周阅读摘选 2026-04-13 → 2026-04-19 目录 学术相关 Towards real-world human behavior simulation: benchmarking large language models on long-horizon, cross-scenario, heterogeneous behavior traces An introduction to double/debiased machine learning...