影响力指数

97.03/100

前 0.1%

全站排名 #95

发表论文71 篇

平均评分5.2

年均产出23.7 篇/年

Feng Liu

Associate Professor@University of Melbourne·澳大利亚·OpenReview

研究方向

Out-of-Distribution Detection · Out-of-Distribution Generalization · Adversatial Machine Learning · Hypothesis testing · Weakly supervised learning · Transfer Learning

Feng Liu

On the Bayes Inconsistency of Disagreement Discrepancy Surrogates

Unlearning Evaluation through Subset Statistical Independence

INRCT: An End-to-End Framework for Encoding and Generating Implicit Neural Representation

POMP: A Theoretical Approach to Mitigate Forgetting in Finetuning Multi-Modal Models

Combating Data Laundering in LLM Training

FedReLa: Imbalanced Federated Learning via Re-Labeling

Beyond Hard Supervised Fine-tuning: Enhancing Image-text Alignment of Strong Models with Weak Models

A Closer Look at In-context Learning of LLMs in Simple Classification Tasks

Solving Puzzles? Jailbreaking Multimodal Large Language Models!

Robust Learning of Diffusion Models with Extremely Noisy Conditions

Editing by Reconstruction: Background Preservation for Instruction-based Autoregressive Image Editing

A Kernel Distribution Closeness Testing

CARPRT: Class-Aware Zero-Shot Prompt Reweighting for Vision-Language Model

RAFT: Reasoning-Aware Fine-Tuning

Improving Generalizability and Undetectability for Targeted Adversarial Attacks on Multimodal Pre-trained Models

Parameter-wise Weighted Model Editing for Efficient and Retentive LLM Unlearning

Once-for-All: Scalable Simultaneous Forecasting via Equilibrium State Estimation

Next-Token Gradient Sensitivity Probing for LLM Hallucination Detection

Prime Once, then Reprogram Locally: An Efficient Alternative to Black-Box Model Reprogramming

Adapting in the Dark: Towards Stable and Efficient Black-Box Test-Time Adaptation

MAPA: Multi-turn Adaptive Prompting Attack On Large Vision-Language Models

Color Blindness Test Images as Seen by Large Vision-Language Models

Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers

DUAL: Learning Diverse Kernels for Aggregated Two-sample and Independence Testing

Generative Model Inversion Through the Lens of the Manifold Hypothesis

Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection

Revealing Multimodal Causality with Large Language Models

Anchor-based Maximum Discrepancy for Relative Similarity Testing

Towards Accurate Time Series Forecasting via Implicit Decoding

Practical Kernel Selection for Kernel-based Conditional Independence Test

Learning Imbalanced Data with Beneficial Label Noise

Understanding Model Reprogramming for CLIP via Decoupling Visual Prompts

CALM: Consensus-Aware Localized Merging for Multi-Task Learning

Long-tailed Recognition with Model Rebalancing

A Closer Look at Backdoor Attacks on CLIP

DDAD: A Two-Pronged Adversarial Defense Based on Distributional Discrepancy

GRU: Mitigating the Trade-off between Unlearning and Retention for LLMs

One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional Discrepancy

Sample-specific Noise Injection for Diffusion-based Adversarial Purification

Test-Time Multimodal Backdoor Detection by Contrastive Prompting

Attribute-based Visual Reprogramming for Vision-Language Models

Deep Kernel Relative Test for Machine-generated Text Detection

CARPRT: Class-Aware Prompt Reweighting for Pre-Trained Vision-Language Models

A Kernel Distribution Closeness Testing

BDetCLIP: Multimodal Prompting Contrastive Test-Time Backdoor Detection

Endowing Visual Reprogramming with Adversarial Robustness

Learning Imbalanced Data with Beneficial Label Noise

Learning Representations for Independence Testing

Sample-specific Noise Injection for Diffusion-based Adversarial Purification

One-step Image-function Generation via Consistency Training

A Closer Look at Backdoor Attacks on CLIP

Privacy Breach Detection by Non-Parametric Two-Sample Tests

Revisit Non-parametric Two-sample Testing as a Semi-supervised Learning Problem

Exploring the Discriminative Capability of LLMs in In-context Learning

Exploiting Presentative Feature Distributions for Parameter-Efficient Continual Learning of Large Language Models

'No' Matters: Out-of-Distribution Detection in Multimodality Long Dialogue

Practical Kernel Learning for Kernel-based Conditional Independent Test

Single-Step Diffusion Model-Based Generative Model Inversion Attacks

Exploring Weak-to-Strong Generalization for CLIP-based Classification