Xinmei Tian

papers

756

total citations

papers (23)

From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning

ICML 2024arXiv

citations

Revisiting Knowledge Distillation: An Inheritance and Exploration Framework

CVPR 2021arXiv

citations

FedImpro: Measuring and Improving Client Update in Federated Learning

ICLR 2024arXiv

citations

Towards Theoretical Understandings of Self-Consuming Generative Models

ICML 2024arXiv

citations

Learning to Localize Actions from Moments

ECCV 2020arXiv

citations

Adversarial Auto-Augment with Label Preservation: A Representation Learning Principle Guided Approach

NEURIPS 2022arXiv

citations

Robust Training of Federated Models with Extremely Label Deficiency

ICLR 2024arXiv

citations

Enhancing Target-unspecific Tasks through a Features Matrix

ICML 2025arXiv

citations

Towards Lightweight Black-Box Attack Against Deep Neural Networks

NEURIPS 2022arXiv

citations

A Similarity Paradigm Through Textual Regularization Without Forgetting

AAAI 2025arXiv

citations

Detecting Generated Images by Fitting Natural Image Distributions

NEURIPS 2025arXiv

citations

Epistemic Uncertainty for Generated Image Detection

NEURIPS 2025arXiv

citations

3D Local Convolutional Neural Networks for Gait Recognition

ICCV 2021

citations

Class-Disentanglement and Applications in Adversarial Detection and Defense

NEURIPS 2021

citations

Meta Convolutional Neural Networks for Single Domain Generalization

CVPR 2022

citations

Enhanced Motion-Text Alignment for Image-to-Video Transfer Learning

CVPR 2024

citations

Learning To Collaborate in Decentralized Learning of Personalized Models

CVPR 2022

citations

Sheared Backpropagation for Fine-tuning Foundation Models

CVPR 2024

citations

Xinmei Tian

papers (23)

Prompt Distribution Learning

FedFed: Feature Distillation against Data Heterogeneity in Federated Learning

Self-Supervision Can Be a Good Few-Shot Learner

A Style and Semantic Memory Mechanism for Domain Generalization

Interpreting and Improving Large Language Models in Arithmetic Calculation

From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning

Revisiting Knowledge Distillation: An Inheritance and Exploration Framework

FedImpro: Measuring and Improving Client Update in Federated Learning

Towards Theoretical Understandings of Self-Consuming Generative Models

Learning to Localize Actions from Moments

Adversarial Auto-Augment with Label Preservation: A Representation Learning Principle Guided Approach

Robust Training of Federated Models with Extremely Label Deficiency

Enhancing Target-unspecific Tasks through a Features Matrix

Towards Lightweight Black-Box Attack Against Deep Neural Networks

A Similarity Paradigm Through Textual Regularization Without Forgetting

Detecting Generated Images by Fitting Natural Image Distributions

Epistemic Uncertainty for Generated Image Detection

3D Local Convolutional Neural Networks for Gait Recognition

Class-Disentanglement and Applications in Adversarial Detection and Defense

Meta Convolutional Neural Networks for Single Domain Generalization

Enhanced Motion-Text Alignment for Image-to-Video Transfer Learning

Learning To Collaborate in Decentralized Learning of Personalized Models

Sheared Backpropagation for Fine-tuning Foundation Models

papers (23)

Prompt Distribution Learning

FedFed: Feature Distillation against Data Heterogeneity in Federated Learning

Self-Supervision Can Be a Good Few-Shot Learner

A Style and Semantic Memory Mechanism for Domain Generalization

Interpreting and Improving Large Language Models in Arithmetic Calculation

From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning

Revisiting Knowledge Distillation: An Inheritance and Exploration Framework

FedImpro: Measuring and Improving Client Update in Federated Learning

Towards Theoretical Understandings of Self-Consuming Generative Models

Learning to Localize Actions from Moments

Adversarial Auto-Augment with Label Preservation: A Representation Learning Principle Guided Approach

Robust Training of Federated Models with Extremely Label Deficiency

Enhancing Target-unspecific Tasks through a Features Matrix

Towards Lightweight Black-Box Attack Against Deep Neural Networks

A Similarity Paradigm Through Textual Regularization Without Forgetting

Detecting Generated Images by Fitting Natural Image Distributions

Epistemic Uncertainty for Generated Image Detection

3D Local Convolutional Neural Networks for Gait Recognition

Class-Disentanglement and Applications in Adversarial Detection and Defense

Meta Convolutional Neural Networks for Single Domain Generalization

Enhanced Motion-Text Alignment for Image-to-Video Transfer Learning

Learning To Collaborate in Decentralized Learning of Personalized Models

Sheared Backpropagation for Fine-tuning Foundation Models