Linghe Kong

IDER: IDempotent Experience Replay for Reliable Continual Learning

Pedagogically-Inspired Data Synthesis for Language Model Knowledge Distillation

Enhanced Continual Learning of Vision-Language Models with Model Fusion

InfVSR: Breaking Length Limits of Generic Video Super-Resolution

GeoFlow: Geo-Aware Modeling of Inter-Area Relationships in OD Flow Prediction and Generation

DVD-Quant: Data-free Video Diffusion Transformers Quantization

CondiQuant: Condition Number Based Low-Bit Quantization for Image Super-Resolution

Revisiting Visual Understanding in Multimodal Reasoning through a Lens of Image Perturbation

ReCalKV: Low-Rank KV Cache Compression via Head Reordering and Offline Calibration

RADAR: A Risk-Aware Dynamic Multi-Agent Framework for LLM Safety Evaluation via Role-Specialized Collaboration

CodeGenGuard: A Watermark for Code Generation Models

Prototype Decoding and Re-Encoding for Class-Incremental Learning

Loss-Aligned Structured Pruning for Large Language Models

AdaSVD: Adaptive Singular Value Decomposition for Large Language Models

CLQ: Cross-Layer Guided Orthogonal-based Quantization for Diffusion Transformers

LLM-IQA: Standard-guided MLLM for Mix-grained Image Quality Assessment

FlashEdit: Decoupling Speed, Structure, and Semantics for Precise Image Editing

RUN: Reversible Unfolding Network for Concealed Object Segmentation

ARB-LLM: Alternating Refined Binarizations for Large Language Models

First SFT, Second RL, Third UPT: Continual Improving Multi-Modal LLM Reasoning via Unsupervised Post-Training

PANTHER: Generative Pretraining Beyond Language for Sequential User Behavior Modeling

CMamba: Channel Correlation Enhanced State Space Models for Multivariate Time Series Forecasting

BiDRN: Binarized 3D Whole-body Human Mesh Recovery

Multi-player Multi-armed Bandits with Delayed Feedback

Dog-IQA: Standard-guided Zero-shot MLLM for Mix-grained Image Quality Assessment

BiMaCoSR: Binary One-Step Diffusion Model Leveraging Flexible Matrix Compression for Real Super-Resolution

LoRA-Switch: Boosting the Efficiency of Dynamic LLM Adapters via System-Algorithm Co-design

Image Super-Resolution with Text Prompt Diffusion

Preserving Large Activations: The Key to KV Cache Pruning

OSM+: Cloud-native Open Street Map Data System for City-wide Experiments