User Behavior Simulation with Large Language Model based Agents
基于认知科学实现推荐用户模拟, 并以此探究信息茧房、从众心理等现象
Empirical
Code Graph Model (CGM): A Graph-Integrated Large Language Model for Repository-Level Software Engineering Tasks
Code Graph & Graph Attention Mask
RepoGraph: Enhancing AI Software Engineering with Repository-level Code Graph
通过 Repository-level 的 Code Graph 实现更为高效的工程代码生成
Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph
在知识图谱上实现 LLM, KG 的交替推理
Learnable Item Tokenization for Generative Recommendation
兼顾语义和协同信息的 Tokenizer
Climber: Toward Efficient Scaling Laws for Large Recommendation Models
一个 CTR 领域更加高效 Scaling 的推荐架构
Pctx: Tokenizing Personalized Context for Generative Recommendation
个性化 semantic IDs
Scaling Laws
Scaling Laws 的细节推导
UniSearch: Rethinking Search System with a Unified Generative Architecture
End-to-End 生成式检索
RecBase: Generative Foundation Model Pretraining for Zero-Shot Recommendation
在多个开源数据集上训练的生成式推荐模型, 利用课程学习训练 Tokenizer