Copertina

No Hard Negatives Required: Concept Centric Learning Leads to Compositionality without Degrading Zero-shot Capabilities of Contrastive Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Pham, Hai X., Hoffmann, David T., Guerrero, Ricardo, Martinez, Brais
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computer Vision and Pattern Recognition Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Unified Framework for Open-World Compositional Zero-shot Learning
di: Jayasekara, Hirunima, et al.
Pubblicazione: (2024)

Interpretable Zero-shot Learning with Infinite Class Concepts
di: Ye, Zihan, et al.
Pubblicazione: (2025)

Visual Perturbation and Adaptive Hard Negative Contrastive Learning for Compositional Reasoning in Vision-Language Models
di: Huang, Xin, et al.
Pubblicazione: (2025)

Locality-Sensitive Hashing for Efficient Hard Negative Sampling in Contrastive Learning
di: Deuser, Fabian, et al.
Pubblicazione: (2025)

FAM Diffusion: Frequency and Attention Modulation for High-Resolution Image Generation with Stable Diffusion
di: Yang, Haosen, et al.
Pubblicazione: (2024)

Zero-shot Object-Centric Instruction Following: Integrating Foundation Models with Traditional Navigation
di: Raychaudhuri, Sonia, et al.
Pubblicazione: (2024)

Enhancing Contrastive Learning for Geolocalization by Discovering Hard Negatives on Semivariograms
di: Chen, Boyi, et al.
Pubblicazione: (2025)

Learning Clustering-based Prototypes for Compositional Zero-shot Learning
di: Qu, Hongyu, et al.
Pubblicazione: (2025)

Contrasting Intra-Modal and Ranking Cross-Modal Hard Negatives to Enhance Visio-Linguistic Compositional Understanding
di: Zhang, Le, et al.
Pubblicazione: (2023)

Explaining CLIP Zero-shot Predictions Through Concepts
di: Ozdemir, Onat, et al.
Pubblicazione: (2026)

A Conditional Probability Framework for Compositional Zero-shot Learning
di: Wu, Peng, et al.
Pubblicazione: (2025)

Zero-shot Concept Bottleneck Models
di: Yamaguchi, Shin'ya, et al.
Pubblicazione: (2025)

Compositional Zero-shot Learning via Progressive Language-based Observations
di: Li, Lin, et al.
Pubblicazione: (2023)

ACE: Concept Editing in Diffusion Models without Performance Degradation
di: Wang, Ruipeng, et al.
Pubblicazione: (2025)

Graph-guided Cross-composition Feature Disentanglement for Compositional Zero-shot Learning
di: Geng, Yuxia, et al.
Pubblicazione: (2024)

SynCo: Synthetic Hard Negatives for Contrastive Visual Representation Learning
di: Giakoumoglou, Nikolaos, et al.
Pubblicazione: (2024)

Conceptrol: Concept Control of Zero-shot Personalized Image Generation
di: He, Qiyuan, et al.
Pubblicazione: (2025)

ActionCOMET: A Zero-shot Approach to Learn Image-specific Commonsense Concepts about Actions
di: Sampat, Shailaja Keyur, et al.
Pubblicazione: (2024)

Contextual Interaction via Primitive-based Adversarial Training For Compositional Zero-shot Learning
di: Li, Suyi, et al.
Pubblicazione: (2024)

Training-free Zero-shot Composed Image Retrieval with Local Concept Reranking
di: Sun, Shitong, et al.
Pubblicazione: (2023)

Compositional Zero-Shot Learning with Contextualized Cues and Adaptive Contrastive Training
di: Li, Yun, et al.
Pubblicazione: (2024)

Zero-shot Compositional Action Recognition with Neural Logic Constraints
di: Ye, Gefan, et al.
Pubblicazione: (2025)

ZeroComp: Zero-shot Object Compositing from Image Intrinsics via Diffusion
di: Zhang, Zitian, et al.
Pubblicazione: (2024)

Contrastive CFG: Improving CFG in Diffusion Models by Contrasting Positive and Negative Concepts
di: Chang, Jinho, et al.
Pubblicazione: (2024)

Sharpness-Aware Data Generation for Zero-shot Quantization
di: Hoang-Anh, Dung, et al.
Pubblicazione: (2025)

Understanding Hyperbolic Metric Learning through Hard Negative Sampling
di: Yue, Yun, et al.
Pubblicazione: (2024)

Rehabilitation Exercise Quality Assessment through Supervised Contrastive Learning with Hard and Soft Negatives
di: Karlov, Mark, et al.
Pubblicazione: (2024)

Enhancing Conceptual Understanding in Multimodal Contrastive Learning through Hard Negative Samples
di: Rösch, Philipp J., et al.
Pubblicazione: (2024)

Do LLMs Understand Visual Anomalies? Uncovering LLM's Capabilities in Zero-shot Anomaly Detection
di: Zhu, Jiaqi, et al.
Pubblicazione: (2024)

ZeroIDIR: Zero-Reference Illumination Degradation Image Restoration with Perturbed Consistency Diffusion Models
di: Jiang, Hai, et al.
Pubblicazione: (2026)

Contrastive Spectral Rectification: Test-Time Defense towards Zero-shot Adversarial Robustness of CLIP
di: Nie, Sen, et al.
Pubblicazione: (2026)

Zero-shot Low-Field MRI Enhancement via Diffusion-Based Adaptive Contrast Transport
di: Liu, Muyu, et al.
Pubblicazione: (2026)

MMGeoLM: Hard Negative Contrastive Learning for Fine-Grained Geometric Understanding in Large Multimodal Models
di: Sun, Kai, et al.
Pubblicazione: (2025)

Prompting without Panic: Attribute-aware, Zero-shot, Test-Time Calibration
di: Hebbalaguppe, Ramya, et al.
Pubblicazione: (2025)

ZeroDiff++: Substantial Unseen Visual-semantic Correlation in Zero-shot Learning
di: Ye, Zihan, et al.
Pubblicazione: (2026)

CLIP-driven Zero-shot Learning with Ambiguous Labels
di: Fan, Jinfu, et al.
Pubblicazione: (2026)

Prioritized Semantic Learning for Zero-shot Instance Navigation
di: Sun, Xinyu, et al.
Pubblicazione: (2024)

Zero-shot Adaptation of Stable Diffusion via Plug-in Hierarchical Degradation Representation for Real-World Super-Resolution
di: Liao, Yi-Cheng, et al.
Pubblicazione: (2025)

Knowledge Distillation Meets Open-Set Semi-Supervised Learning
di: Yang, Jing, et al.
Pubblicazione: (2022)

Gate-and-Merge: Zero-shot Compositional Personalization of Vision Language Models
di: Ding, Guodong, et al.
Pubblicazione: (2026)