Bin Chen

Full Professor@Harbin Institute of Technology, Shenzhen·中国·OpenReview

研究方向

Information Theory · Multimedia Compression · AI Safety · Deep Learning · Image Compression · Information Retrieval · Data Compression · Machine Learning · Algebra and its application · Coding theory

Bin Chen

PromptHub: Enhancing Multi-Prompt Visual In-Context Learning with Locality-Aware Fusion, Concentration and Alignment

Closing the Safety Gap: Surgical Concept Erasure in Visual Autoregressive Models

Autoregressive-based Progressive Coding for Ultra-Low Bitrate Image Compression

CoPRS: Learning Positional Prior from Chain-of-Thought for Reasoning Segmentation

Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs

PIC: Revisiting INR for Image Coding with Fast Encoding and Sub-Millisecond Decoding

Rectified Decoupled Dataset Distillation: A Closer Look for Fair and Comprehensive Evaluation

Temporal Saliency-Guided Distillation: A Scalable Framework for Distilling Video Datasets

MambaSIC: Mamba-based Stereo Image Compression with Bi-directional Multi-reference Entropy Model

Talk Until You Burn Out: Escalating 3D-LLM Overgeneration via Semantic Manipulation

Grounding Language with Vision: A Conditional Mutual Information Calibrated Decoding Strategy for Reducing Hallucinations in LVLMs

3D-LMVIC: Learning-based Multi-View Image Compression with 3D Gaussian Geometric Priors

An Exploration with Entropy Constrained 3D Gaussians for 2D Video Compression

Clients Collaborate: Flexible Differentially Private Federated Learning with Guaranteed Improvement of Utility-Privacy Trade-off

Stealthy Shield Defense: A Conditional Mutual Information-Based Approach against Black-Box Model Inversion Attacks

DiffPC: Diffusion-based High Perceptual Fidelity Image Compression with Semantic Refinement

3D-GP-LMVIC: Learning-based Multi-View Image Compression with 3D Gaussian Geometric Priors

Going Beyond Feature Similarity: Effective Dataset distillation based on Class-aware Conditional Mutual Information

Block-to-Scene Pre-training for Point Cloud Hybrid-Domain Masked Autoencoders

One Perturbation is Enough: On Generating Universal Adversarial Perturbations against Vision-Language Pre-training Models

Towards Scalable Semantic Representation for Recommendation

MambaVC: Exploring Selective State Spaces for Learned Visual Compression

CALoR: Towards Comprehensive Model Inversion Defense

MIBench: A Comprehensive Benchmark for Model Inversion Attack and Defense

WATER-GS: Watermark-embedded 3D Gaussian Splatting via Plug-and-play Decoder

RouGE: Learning Gated Experts for Segment Anything in the Wild