Xiaowen Chu

Ghost in the Cloud: Your Geo-Distributed Large Language Models Training is Easily Manipulated

Towards Universal Video Retrieval: Generalizing Video Embedding via Synthesized Multimodal Pyramid Curriculum

Reasoning Language Model Inference Serving Unveiled: An Empirical Study

DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference

Sliding Window Attention for Reinforced Reasoning

OracleKV: Oracle Guidance for Question-Independent KV Cache Eviction

JailbreakLoRA: Your Downloaded LoRA from Sharing Platforms might be Unsafe

AnTKV: Anchor Token-Aware Ultra-Low-Bit Vector Quantization for KV Cache in Large Language Models

Mediater: Memory-efficient LLM Merging with Less Parameter Conflicts and Uncertainty Based Routing

Is Privacy Always Prioritized over Learning? Probing LLMs' Value Priority Belief under External Perturbations

LoRA-FA: Efficient and Effective Low Rank Representation Fine-tuning

FlowKV: Enhancing Multi-Turn Conversational Coherence in LLMs via Isolated Key-Value Cache Management

Identifying and Mitigating Errors in Gradient Aggregation of Distributed Data Parallel Training

Can LLMs Maintain Fundamental Abilities under KV Cache Compression?

Can Compressed LLMs Truly Act? An Empirical Evaluation of Agentic Capabilities in LLM Compression

ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference

Hot-pluggable Federated Learning: Bridging General and Personalized FL via Dynamic Selection

STBLLM: Breaking the 1-Bit Barrier with Structured Binary LLMs

ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference

Capturing and Mitigating Gradient Aggregation Errors for Fault-Tolerant Distributed Training