Xu Chen

CurES: From Gradient Analysis to Efficient Curriculum Learning for Reasoning LLMs

VeriRole: Verifiable Role-Awareness through Hint-Guided Reinforcement Learning

Prompt and Parameter Co-Optimization for Large Language Models

Benchmarking and Enhancing Rational Preference Utilization for Personalized Assistants: A Pragmatic View

Solving the Granularity Mismatch: Hierarchical Preference Learning for Long-Horizon LLM Agents

Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding

Estimating the Effects of Sample Training Orders for Large Language Models without Retraining

Cog-Rethinker: Hierarchical Metacognitive Reinforcement Learning for LLM Reasoning

Mitigating Hidden Confounding by Progressive Confounder Imputation via Large Language Models

Learn to Memorize: Optimizing LLM-based Agents with Adaptive Memory Framework

Is It Necessary to Inject Causality into Chain-of-Thought Reasoning?

Time-o1: Time-Series Forecasting Needs Transformed Label Alignment

MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework

Distributional LLM-as-a-Judge

Uncertainty and Influence aware Reward Model Refinement for Reinforcement Learning from Human Feedback

Optimal Transport for Time Series Imputation

Inverse Methods for Missing Data Imputation

CAM: A Constructivist View of Agentic Memory for LLM-Based Reading Comprehension

Unbiased Recommender Learning from Implicit Feedback via Weakly Supervised Learning

MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents

MemSim: A Bayesian Simulator for Evaluating Memory of LLM-based Personal Assistants

KnowTrace: Explicit Knowledge Tracing for Structured Retrieval-Augmented Generation

Attaining Human's Desirable Outcomes in Indirect Human-AI Interaction via Multi-Agent Influence Diagrams

MemSim: A Bayesian Simulator for Evaluating Memory of LLM-based Personal Assistants

Robust Heterogeneous Treatment Effect Estimation under Covariate Perturbation

Sampling Process Brings Additional Bias for Debiased Recommendation

Mitigating Unobserved Confounding via Diffusion Probabilistic Models

Active Causal Learning for Conditional Average Treatment Effect Estimation