Image de couverture de livre

Hard Labels In! Rethinking the Role of Hard Labels in Mitigating Local Semantic Drift

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Cui, Jiacheng, Tong, Bingkui, Bi, Xinyue, Zhao, Xiaohan, Liu, Jiacheng, Shen, Zhiqiang
Format:	Preprint
Publié:	2025
Sujets:	Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

FADRM: Fast and Accurate Data Residual Matching for Dataset Distillation
par: Cui, Jiacheng, et autres
Publié: (2025)

Rethinking Dataset Distillation: Hard Truths about Soft Labels
par: Dey, Priyam, et autres
Publié: (2026)

Exploring 3D Dataset Pruning
par: Zhao, Xiaohan, et autres
Publié: (2026)

Open CaptchaWorld: A Comprehensive Web-based Platform for Testing and Benchmarking Multimodal LLM Agents
par: Luo, Yaxin, et autres
Publié: (2025)

Dataset Distillation via Committee Voting
par: Cui, Jiacheng, et autres
Publié: (2025)

Mitigating Hallucination in Multimodal LLMs with Layer Contrastive Decoding
par: Tong, Bingkui, et autres
Publié: (2025)

BiGain: Unified Token Compression for Joint Generation and Classification
par: Liu, Jiacheng, et autres
Publié: (2026)

Hard to See, Hard to Label: Generative and Symbolic Acquisition for Subtle Visual Phenomena
par: Prasad, Renjith, et autres
Publié: (2026)

Pushing the Frontier of Black-Box LVLM Attacks via Fine-Grained Detail Targeting
par: Zhao, Xiaohan, et autres
Publié: (2026)

Hard-Label Black-Box Attacks on 3D Point Clouds
par: Liu, Daizong, et autres
Publié: (2024)

OccLE: Label-Efficient 3D Semantic Occupancy Prediction
par: Fang, Naiyu, et autres
Publié: (2025)

A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1
par: Li, Zhaoyi, et autres
Publié: (2025)

Learning with Instance-Dependent Noisy Labels by Anchor Hallucination and Hard Sample Label Correction
par: Huang, Po-Hsuan, et autres
Publié: (2024)

CUE: Concept-Aware Multi-Label Expansion to Mitigate Concept Confusion in Long-Tailed Learning
par: Zhang, Ruichi, et autres
Publié: (2026)

Query-Efficient Hard-Label Black-Box Attack against Vision Transformers
par: Zhou, Chao, et autres
Publié: (2024)

HACMatch Semi-Supervised Rotation Regression with Hardness-Aware Curriculum Pseudo Labeling
par: Li, Mei, et autres
Publié: (2026)

Using Unreliable Pseudo-Labels for Label-Efficient Semantic Segmentation
par: Wang, Haochen, et autres
Publié: (2023)

Towards Eliminating Hard Label Constraints in Gradient Inversion Attacks
par: Wang, Yanbo, et autres
Publié: (2024)

DiN: Diffusion Model for Robust Medical VQA with Semantic Noisy Labels
par: Guo, Erjian, et autres
Publié: (2025)

Accelerating Targeted Hard-Label Adversarial Attacks in Low-Query Black-Box Settings
par: Swaminathan, Arjhun, et autres
Publié: (2025)

Federated Self-supervised Domain Generalization for Label-efficient Polyp Segmentation
par: Tan, Xinyi, et autres
Publié: (2025)

Leveraging Semi-Supervised Learning to Enhance Data Mining for Image Classification under Limited Labeled Data
par: Shen, Aoran, et autres
Publié: (2024)

IPSeg: Image Posterior Mitigates Semantic Drift in Class-Incremental Segmentation
par: Yu, Xiao, et autres
Publié: (2025)

Towards Efficient Pixel Labeling for Industrial Anomaly Detection and Localization
par: Li, Hanxi, et autres
Publié: (2024)

Measuring Epistemic Humility in Multimodal Large Language Models
par: Tong, Bingkui, et autres
Publié: (2025)

Love Me, Love My Label: Rethinking the Role of Labels in Prompt Retrieval for Visual In-Context Learning
par: Luo, Tianci, et autres
Publié: (2026)

Incomplete Multi-Label Image Recognition by Co-learning Semantic-Aware Features and Label Recovery
par: He, Zhi-Fen, et autres
Publié: (2025)

SAVAA: Mitigating Hallucinations in LVLMs via Step-wise Adaptive Visual Attention Amplification
par: Zhang, Jiacheng, et autres
Publié: (2026)

Learning Semantic-Aware Threshold for Multi-Label Image Recognition with Partial Labels
par: Ruan, Haoxian, et autres
Publié: (2025)

LEMoN: Label Error Detection using Multimodal Neighbors
par: Zhang, Haoran, et autres
Publié: (2024)

Enhanced Sample Selection with Confidence Tracking: Identifying Correctly Labeled yet Hard-to-Learn Samples in Noisy Data
par: Pan, Weiran, et autres
Publié: (2025)

Right Regions, Wrong Labels: Semantic Label Flips in Segmentation under Correlation Shift
par: Achara, Akshit, et autres
Publié: (2026)

FerKD: Surgical Label Adaptation for Efficient Distillation
par: Shen, Zhiqiang
Publié: (2023)

SRL-CLIP: Efficient CLIP Video Adaptation via Structured Semantic Role Labels
par: Singh, Darshan, et autres
Publié: (2024)

Multi-graph Graph Matching for Coronary Artery Semantic Labeling
par: Zhao, Chen, et autres
Publié: (2024)

Pushing Rendering Boundaries: Hard Gaussian Splatting
par: Xu, Qingshan, et autres
Publié: (2024)

Hard-aware Instance Adaptive Self-training for Unsupervised Cross-domain Semantic Segmentation
par: Zhu, Chuang, et autres
Publié: (2023)

Category-wise Fine-Tuning: Resisting Incorrect Pseudo-Labels in Multi-Label Image Classification with Partial Labels
par: Chong, Chak Fong, et autres
Publié: (2024)

Not All Voxels Are Equal: Hardness-Aware Semantic Scene Completion with Self-Distillation
par: Wang, Song, et autres
Publié: (2024)

Learning Semantic-Aware Representation in Visual-Language Models for Multi-Label Recognition with Partial Labels
par: Ruan, Haoxian, et autres
Publié: (2024)