Kun Kuang

28

Papers

52

Total Citations

Papers (28)

Towards Better Alignment: Training Diffusion Models with Reinforcement Learning Against Sparse Rewards

MetaCoCo: A New Few-Shot Classification Benchmark with Spurious Correlation

MergeNet: Knowledge Migration Across Heterogeneous Models, Tasks, and Modalities

AuG-KD: Anchor-Based Mixup Generation for Out-of-Domain Knowledge Distillation

Decoding Correlation-Induced Misalignment in the Stable Diffusion Workflow for Text-to-Image Generation

Arrow: Accelerator for Time Series Causal Discovery with Time Weaving

Optimize Incompatible Parameters Through Compatibility-aware Knowledge Integration

Distributionally Generative Augmentation for Fair Facial Attribute Classification

A Generative Approach for Treatment Effect Estimation under Collider Bias: From an Out-of-Distribution Perspective

Learning Shadow Variable Representation for Treatment Effect Estimation under Collider Bias

Two-Stage Shadow Inclusion Estimation: An IV Approach for Causal Inference under Latent Confounding and Collider Bias

Learning Causal Relations from Subsampled Time Series with Two Time-Slices

Model Tailor: Mitigating Catastrophic Forgetting in Multi-modal Large Language Models

Semi-Supervised Active Learning for Semi-Supervised Models: Exploit Adversarial Examples With Graph-Based Virtual Labels

Universal Domain Adaptation via Compressive Attention Matching

MAP: Towards Balanced Generalization of IID and OOD through Model-Agnostic Adapters

InfiAgent-DABench: Evaluating Agents on Data Analysis Tasks

FedCFA: Alleviating Simpson’s Paradox in Model Aggregation with Counterfactual Federated Learning

Learning Causal Transition Matrix for Instance-dependent Label Noise

De-biased Attention Supervision for Text Classification with Causality

Learning to Reweight for Generalizable Graph Neural Network

CGMGM: A Cross-Gaussian Mixture Generative Model for Few-Shot Semantic Segmentation

Contrastive Balancing Representation Learning for Heterogeneous Dose-Response Curves Estimation

GRASP: Navigating Retrosynthetic Planning with Goal-driven Policy

ConfounderGAN: Protecting Image Data Privacy with Causal Confounder

Two Heads are Better Than One: A Simple Exploration Framework for Efficient Multi-Agent Reinforcement Learning

HAP: Structure-Aware Masked Image Modeling for Human-Centric Perception

Disentangled Graph Convolutional Networks