Hongkai Xiong

25

Papers

12

Total Citations

Papers (25)

Betrayed by Attention: A Simple yet Effective Approach for Self-supervised Video Object Segmentation

METEOR: Multi-Encoder Collaborative Token Pruning for Efficient Vision Language Models

HiPART: Hierarchical Pose AutoRegressive Transformer for Occluded 3D Human Pose Estimation

Diffusion-Driven Progressive Target Manipulation for Source-Free Domain Adaptation

NeurIPS 2025arXiv

LBI-FL: Low-Bit Integerized Federated Learning with Temporally Dynamic Bit-Width Allocation

Improving Diffusion Models for Inverse Problems Using Optimal Posterior Covariance

Bootstrap AutoEncoders With Contrastive Paradigm for Self-supervised Gaze Estimation

AMPA: Adaptive Mixed Precision Allocation for Low-Bit Integer Training

Picking Deep Filter Responses for Fine-Grained Image Recognition

Zigzag Learning for Weakly Supervised Object Detection

Rotation Equivariant Graph Convolutional Network for Spherical Image Classification

Motion-Aware Contrastive Video Representation Learning via Foreground-Background Merging

Contrastive Regression for Domain Adaptation on Gaze Estimation

Adapting Shortcut With Normalizing Flow: An Efficient Tuning Framework for Visual Recognition

Promoting Semantic Connectivity: Dual Nearest Neighbors Contrastive Learning for Unsupervised Domain Generalization

Learning Latent Architectural Distribution in Differentiable Neural Architecture Search via Variational Information Maximization

Towards Unsupervised Domain Generalization for Face Anti-Spoofing

Prune Spatio-temporal Tokens by Semantic-aware Temporal Accumulation

Source-Free Domain Adaptation with Contrastive Domain Alignment and Self-Supervised Exploration for Face Anti-Spoofing

A Transformer-Based Decoder for Semantic Segmentation with Multi-level Context Mining

SdAE: Self-Distillated Masked Autoencoder

IM-Zero: Instance-level Motion Controllable Video Generation in a Zero-shot Manner

Stabilizing and Accelerating Autofocus with Expert Trajectory Regularized Deep Reinforcement Learning

AiluRus: A Scalable ViT Framework for Dense Prediction

Doubly Robust Augmented Transfer for Meta-Reinforcement Learning