Xue Liu

Escaping Policy Contraction: Contraction-Aware PPO (CaPPO) for Stable Language Model Fine-Tuning

Spatial CAPTCHA: Generatively Benchmarking Spatial Reasoning for Human-Machine Differentiation

Pedagogically-Inspired Data Synthesis for Language Model Knowledge Distillation

Adversarial Encoding Perturbation and Synthesis for Set Representation Auxiliary Learning

Tequila: Trapping-free Ternary Quantization for Large Language Models

RAEE: A Robust Retrieval-Augmented Early Exit Framework for Efficient Inference

Prima.cpp: Fast 30-70B LLM Inference on Heterogeneous and Low-Resource Home Clusters

Which Heads Matter for Reasoning? RL-Guided KV Cache Compression

Beyond Semantic Similarity: Reducing Unnecessary API Calls via Behavior-Aligned Retriever

Exploring the Trade-off between Quality and Diversity of Language Models during Reinforcement Learning

RECODE-H: A Benchmark for Research Code Development with Interactive Human Feedback

Mixture of Experts Characteristic Function Embeddings for Heterogeneous Fraud Graphs

Optimizing User Profiles via Contextual Bandits for Retrieval-Augmented LLM Personalization

Light-Search: Reducing Retrieval Cost in RAG via Curriculum-Based Policy Optimization

On Why Form Shapes Reasoning: Structuring Latent Program Networks with Category-Theoretic Constraints

Reasoning at the Right Length: Adaptive Budget Forcing for Efficient and Accurate LLM Inference

Diffusion Large Language Models for Black-Box Optimization

Recall-First Moderation via Distribution-Preserving Augmentation and Committee-Diverse Retrieval

How to Train Your LLM Web Agent: A Statistical Diagnosis

3DMolFormer: A Dual-channel Framework for Structure-based Drug Discovery

ParetoFlow: Guided Flows in Multi-Objective Optimization

RAEE: A Robust Retrieval-Augmented Early Exit Framework for Efficient Inference

T-Graphormer: Using Transformers for Spatiotemporal Forecasting