34 citations · 130 across the 53 of their papers we have counts for
57 papers
Semantic Mismapping in Generalized Category Discovery: Evaluation and Mitigation
Zitai Wang, Qianqian Xu, Cong Hua +4
Generalized Category Discovery (GCD) aims to perform old-category recognition and novel-category discovery in the same unlabeled set. Early methods address both objectives through…
Pairwise AUC Optimization Needs Corrective Power: A Unified View
Jia Chen, Zhiyong Yang, Shilong Bao +2
Direct pairwise AUC-surrogate training is attractive because it trains toward the ranking metric used at test time, especially under class imbalance. This creates a direct alignmen…
Compare to Correct: Pairwise-Guided Logit Adjustment for LLM Test-Time Scaling
Cong Hua, Qianqian Xu, Zitai Wang +4
Sampling-based test-time scaling enhances LLM reasoning by aggregating multiple reasoning paths for the same query. Existing methods use answer frequencies or path confidence to co…
GoodDiffusion: Proactive Copyright Protection for Diffusion Bridge Models via Learnable Sample-specific Signatures
Shixi Qin, Zhiyong Yang, Shilong Bao +3
This paper tackles the challenging problem of developing a proactive copyright protection mechanism that cuts off unauthorized use of diffusion bridge models. Existing studies larg…
Understanding-Enhanced Model Collaboration for Long-Tailed Egocentric Mistake Detection
Boyu Han, Qianqian Xu, Shilong Bao +3
In this report, we address the problem of determining whether a user performs an action incorrectly from egocentric video data. To this end, we propose an Understanding-Enhanced Mo…
The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works
Guanghui Wang, Kaiwen Lv Kacuila, Zhiyong Yang +5
Knowledge distillation (KD) transfers knowledge from a large teacher model to a smaller student. In language modeling, the student is trained either on tokens sampled from the teac…
The Devil is in the Condition Numbers: Why is GLU Better than non-GLU Structure?
Xingyu Lyu, Qianqian Xu, Zhiyong Yang +2
Gated Linear Units (GLU) and their variants are widely adopted in modern open-source large language model architectures and consistently outperform their non-gated counterparts, ye…
Mind the Way You Select Negative Texts: Pursuing the Distance Consistency in OOD Detection with VLMs
Zhikang Xu, Qianqian Xu, Zitai Wang +4
Out-of-distribution (OOD) detection seeks to identify samples from unknown classes, a critical capability for deploying machine learning models in open-world scenarios. Recent rese…
Guiding Diffusion-based Reconstruction with Contrastive Signals for Balanced Visual Representation
Boyu Han, Qianqian Xu, Shilong Bao +4
The limited understanding capacity of the visual encoder in Contrastive Language-Image Pre-training (CLIP) has become a key bottleneck for downstream performance. This capacity inc…
BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation
Feiran Li, Qianqian Xu, Shilong Bao +4
This paper investigates the challenging task of detecting backdoored text-to-image models under black-box settings and introduces a novel detection framework BlackMirror. Existing…
Closing the Approximation Gap of Partial AUC Optimization: A Tale of Two Formulations
Yangbangyan Jiang, Qianqian Xu, Huiyang Shao +4
As a variant of the Area Under the ROC Curve (AUC), the partial AUC (PAUC) focuses on a specific range of false positive rate (FPR) and/or true positive rate (TPR) in the ROC curve…
Quantifying the Potential to Escape Filter Bubbles: A Behavior-Aware Measure via Contrastive Simulation
Difu Feng, Qianqian Xu, Zitai Wang +3
Nowadays, recommendation systems have become crucial to online platforms, shaping user exposure by accurate preference modeling. However, such an exposure strategy can also reinfor…
TuckA: Hierarchical Compact Tensor Experts for Efficient Fine-Tuning
Qifeng Lei, Zhiyong Yang, Qianqian Xu +3
Efficiently fine-tuning pre-trained models for downstream tasks is a key challenge in the era of foundation models. Parameter-efficient fine-tuning (PEFT) presents a promising solu…
Towards Size-invariant Salient Object Detection: A Generic Evaluation and Optimization Approach
Shilong Bao, Qianqian Xu, Feiran Li +4
This paper investigates a fundamental yet underexplored issue in Salient Object Detection (SOD): the size-invariant property for evaluation protocols, particularly in scenarios whe…
Dual-Stage Reweighted MoE for Long-Tailed Egocentric Mistake Detection
Boyu Han, Qianqian Xu, Shilong Bao +3
In this report, we address the problem of determining whether a user performs an action incorrectly from egocentric video data. To handle the challenges posed by subtle and infrequ…
LightFair: Towards an Efficient Alternative for Fair T2I Diffusion via Debiasing Pre-trained Text Encoders
Boyu Han, Qianqian Xu, Shilong Bao +3
This paper explores a novel lightweight approach LightFair to achieve fair text-to-image diffusion models (T2I DMs) by addressing the adverse effects of the text encoder. Most exis…
Hybrid Generative Fusion for Efficient and Privacy-Preserving Face Recognition Dataset Generation
Feiran Li, Qianqian Xu, Shilong Bao +3
In this paper, we present our approach to the DataCV ICCV Challenge, which centers on building a high-quality face dataset to train a face recognition model. The constructed datase…
ABKD: Pursuing a Proper Allocation of the Probability Mass in Knowledge Distillation via --Divergence
Guanghui Wang, Zhiyong Yang, Zitai Wang +3
Knowledge Distillation (KD) transfers knowledge from a large teacher model to a smaller student model by minimizing the divergence between their output distributions, typically usi…
Focal-SAM: Focal Sharpness-Aware Minimization for Long-Tailed Classification
Sicong Li, Qianqian Xu, Zhiyong Yang +4
Real-world datasets often follow a long-tailed distribution, making generalization to tail classes difficult. Recent methods resorted to long-tail variants of Sharpness-Aware Minim…
One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Framework
Feiran Li, Qianqian Xu, Shilong Bao +3
Concept erasing has recently emerged as an effective paradigm to prevent text-to-image diffusion models from generating visually undesirable or even harmful content. However, curre…
MixBridge: Heterogeneous Image-to-Image Backdoor Attack through Mixture of Schrödinger Bridges
Shixi Qin, Zhiyong Yang, Shilong Bao +3
This paper focuses on implanting multiple heterogeneous backdoor triggers in bridge-based diffusion models designed for complex and arbitrary input distributions. Existing backdoor…
OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning
Cong Hua, Qianqian Xu, Zhiyong Yang +3
Prompt tuning adapts Vision-Language Models like CLIP to open-world tasks with minimal training costs. In this direction, one typical paradigm evaluates model performance separatel…
EDGE: Unknown-aware Multi-label Learning by Energy Distribution Gap Expansion
Yuchen Sun, Qianqian Xu, Zitai Wang +2
Multi-label Out-Of-Distribution (OOD) detection aims to discriminate the OOD samples from the multi-label In-Distribution (ID) ones. Compared with its multiclass counterpart, it is…
SSE-SAM: Balancing Head and Tail Classes Gradually through Stage-Wise SAM
Xingyu Lyu, Qianqian Xu, Zhiyong Yang +2
Real-world datasets often exhibit a long-tailed distribution, where vast majority of classes known as tail classes have only few samples. Traditional methods tend to overfit on the…