Ji-Rong Wen

This paper examines new strategies for enhancing document retrieval through token pruning.

EulerFormer combines semantic and positional information for better user behavior predictions.

An overview of how memory enhances the performance of LLM-based agents.

This article discusses extending context windows in language models using positional vectors.

A new model improves user interest predictions for video content.

Explores the challenges of supervising advanced AI models with weaker counterparts.

Research highlights in-context learning abilities in large language models.

New techniques improve search engine models by considering user context.

A new method improves the search for mathematical expressions from data.

Introducing MemSim, a tool for assessing memory effectiveness in language model assistants.

TrendSim simulates social media trends to study misinformation effects.

New benchmark OmniEval enhances evaluation of RAG systems in finance.