103 citations · 556 across the 52 of their papers we have counts for
33 papers · 1 filter
Channel-wise Dynamic Knowledge Distillation via Adaptive Sample Generation for Action Recognition
Ping Li, Chenhao Ping, Jie Song +1
Knowledge Distillation (KD) offers a promising yet underexplored path for compressing large action recognition models. However, existing KD methods suffer from two key limitations:…
RS3DBench: A Comprehensive Benchmark for 3D Spatial Perception in Remote Sensing
Jiayu Wang, Ruizhi Wang, Jie Song +4
In this paper, we introduce a novel benchmark designed to propel the advancement of general-purpose, large-scale 3D vision models for remote sensing imagery. While several datasets…
Training Data Provenance Verification: Did Your Model Use Synthetic Data from My Generative Model for Training?
Yuechen Xie, Jie Song, Huiqiong Wang +1
High-quality open-source text-to-image models have lowered the threshold for obtaining photorealistic images significantly, but also face potential risks of misuse. Specifically, s…
On the Evaluation Consistency of Attribution-based Explanations
Jiarui Duan, Haoling Li, Haofei Zhang +5
Attribution-based explanations are garnering increasing attention recently and have emerged as the predominant approach towards \textit{eXplanable Artificial Intelligence}~(XAI). H…
A Large-scale Universal Evaluation Benchmark For Face Forgery Detection
Yijun Bei, Hengrui Lou, Jinsong Geng +5
With the rapid development of AI-generated content (AIGC) technology, the production of realistic fake facial images and videos that deceive human visual perception has become poss…
On the Concept Trustworthiness in Concept Bottleneck Models
Qihan Huang, Jie Song, Jingwen Hu +3
Concept Bottleneck Models (CBMs), which break down the reasoning process into the input-to-concept mapping and the concept-to-label prediction, have garnered significant attention…
Training-Free Pretrained Model Merging
Zhengqi Xu, Ke Yuan, Huiqiong Wang +3
Recently, model merging techniques have surfaced as a solution to combine multiple single-talent models into a single multi-talent model. However, previous endeavors in this field…
Lookaround Optimizer: steps around, 1 step average
Jiangtao Zhang, Shunyu Liu, Jie Song +3
Weight Average (WA) is an active research topic due to its simplicity in ensembling deep networks and the effectiveness in promoting generalization. Existing weight average approac…
Generalization Matters: Loss Minima Flattening via Parameter Hybridization for Efficient Online Knowledge Distillation
Tianli Zhang, Mengqi Xue, Jiangtao Zhang +5
Most existing online knowledge distillation(OKD) techniques typically require sophisticated modules to produce diverse knowledge for improving students' generalization ability. In…
Schema Inference for Interpretable Image Classification
Haofei Zhang, Mengqi Xue, Xiaokang Liu +3
In this paper, we study a novel inference paradigm, termed as schema inference, that learns to deductively infer the explainable predictions by rebuilding the prior deep neural net…
Evaluation and Improvement of Interpretability for Self-Explainable Part-Prototype Networks
Qihan Huang, Mengqi Xue, Wenqi Huang +4
Part-prototype networks (e.g., ProtoPNet, ProtoTree, and ProtoPool) have attracted broad research interest for their intrinsic interpretability and comparable accuracy to non-inter…
Attention Diversification for Domain Generalization
Rang Meng, Xianfeng Li, Weijie Chen +7
Convolutional neural networks (CNNs) have demonstrated gratifying results at learning discriminative features. However, when applied to unseen domains, state-of-the-art models are…
ProtoPFormer: Concentrating on Prototypical Parts in Vision Transformers for Interpretable Image Recognition
Mengqi Xue, Qihan Huang, Haofei Zhang +4
Prototypical part network (ProtoPNet) has drawn wide attention and boosted many follow-up studies due to its self-explanatory property for explainable artificial intelligence (XAI)…
Federated Selective Aggregation for Knowledge Amalgamation
Donglin Xie, Ruonan Yu, Gongfan Fang +5
In this paper, we explore a new knowledge-amalgamation problem, termed Federated Selective Aggregation (FedSA). The goal of FedSA is to train a student model for a new task with th…
Learning with Recoverable Forgetting
Jingwen Ye, Yifang Fu, Jie Song +5
Life-long learning aims at learning a sequence of tasks without forgetting the previously acquired knowledge. However, the involved training data may not be life-long legitimate du…
Slimmable Domain Adaptation
Rang Meng, Weijie Chen, Shicai Yang +7
Vanilla unsupervised domain adaptation methods tend to optimize the model with fixed neural architecture, which is not very practical in real-world scenarios since the target data…
Label Matching Semi-Supervised Object Detection
Binbin Chen, Weijie Chen, Shicai Yang +6
Semi-supervised object detection has made significant progress with the development of mean teacher driven self-training. Despite the promising results, the label mismatch problem…
Learning Domain Adaptive Object Detection with Probabilistic Teacher
Meilin Chen, Weijie Chen, Shicai Yang +8
Self-training for unsupervised domain adaptive object detection is a challenging task, of which the performance depends heavily on the quality of pseudo boxes. Despite the promisin…
Spot-adaptive Knowledge Distillation
Jie Song, Ying Chen, Jingwen Ye +1
Knowledge distillation (KD) has become a well established paradigm for compressing deep neural networks. The typical way of conducting knowledge distillation is to train the studen…
Meta-attention for ViT-backed Continual Learning
Mengqi Xue, Haofei Zhang, Jie Song +1
Continual learning is a longstanding research topic due to its crucial role in tackling continually arriving tasks. Up to now, the study of continual learning in computer vision is…
Knowledge Amalgamation for Object Detection with Transformers
Haofei Zhang, Feng Mao, Mengqi Xue +4
Knowledge amalgamation (KA) is a novel deep model reusing task aiming to transfer knowledge from several well-trained teachers to a multi-talented and compact student. Currently, m…
Model Doctor: A Simple Gradient Aggregation Strategy for Diagnosing and Treating CNN Classifiers
Zunlei Feng, Jiacong Hu, Sai Wu +3
Recently, Convolutional Neural Network (CNN) has achieved excellent performance in the classification task. It is widely known that CNN is deemed as a 'black-box', which is hard fo…
KDExplainer: A Task-oriented Attention Model for Explaining Knowledge Distillation
Mengqi Xue, Jie Song, Xinchao Wang +3
Knowledge distillation (KD) has recently emerged as an efficacious scheme for learning compact deep neural networks (DNNs). Despite the promising results achieved, the rationale th…
DMN4: Few-shot Learning via Discriminative Mutual Nearest Neighbor Neural Network
Yang Liu, Tu Zheng, Jie Song +2
Few-shot learning (FSL) aims to classify images under low-data regimes, where the conventional pooled global feature is likely to lose useful local characteristics. Recent work has…