Research archive

Publications.

Selected work spanning agent systems, large language models, multimodal intelligence, search, and recommender systems.

17 selected papers
01

χ-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?

Agents

NeurIPS E&D2026
02

BehaviorBench: Modeling Real-World User Decisions from Behavioral Traces

Agents

NeurIPS E&D2026
03

Test-Time Adaptation for LLM Agents via Environment Interaction

Agents

ICLR2026
04

WebScale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining Levels

Agents

ICLR2026
05

UserBench: An Interactive Gym Environment for User-Centric Agents

Agents

EMNLP2026
06

UserRL: A Gym-Based Testbed for User-Centric Reinforcement Learning

Agents

EMNLP2026
07

MCPEval: Automatic MCP-based Deep Evaluation for AI Agent Models

Agents

arXiv2025
08

LATTE: Learning to Think with Vision Specialists

Multimodal

EMNLP2025
09

PRACT: Optimizing Principled Reasoning and Acting of LLM Agent

Agents

CoNLL2024
10

xLAM: A Family of Large Action Models to Empower AI Agent Systems

Agents

arXiv2024
11

AgentLite: A Lightweight Library for Building and Advancing Task-Oriented LLM Agent System

Agents

arXiv2024
12

BOLAA: Benchmarking and Orchestrating LLM-augmented Autonomous Agents

Agents

ICLR LLMAgent2024
13

Retroformer: Retrospective Large Language Agents with Policy Gradient Optimization

Agents

ICLR2024
14

DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AI

Language

arXiv2023
15

Contrastive Self-supervised Sequential Recommendation with Robust Augmentation

Recommendation

Sequential recommendation2021
16

Continuous-Time Sequential Recommendation with Temporal Graph Collaborative Transformer

Recommendation

CIKM2021
17

Augmenting Sequential Recommendation with Pseudo-Prior Items via Reversely Pre-training Transformer

Recommendation

SIGIR2021

Looking for the complete citation record?

View Google Scholar ↗