Imagen de Portada

Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Bhalla, Usha, Oesterling, Alex, Srinivas, Suraj, Calmon, Flavio P., Lakkaraju, Himabindu
Formato:	Preprint
Publicado:	2024
Materias:	Machine Learning Computer Vision and Pattern Recognition
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Discriminative Feature Attributions: Bridging Post Hoc Explainability and Inherent Interpretability
por: Bhalla, Usha, et al.
Publicado: (2023)

Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability
por: Bhalla, Usha, et al.
Publicado: (2025)

All Roads Lead to Rome? Exploring Representational Similarities Between Latent Spaces of Generative Image Models
por: Badrinath, Charumathi, et al.
Publicado: (2024)

Evaluating Adversarial Robustness of Concept Representations in Sparse Autoencoders
por: Li, Aaron J., et al.
Publicado: (2025)

Towards Unifying Interpretability and Control: Evaluation via Intervention
por: Bhalla, Usha, et al.
Publicado: (2024)

Operationalizing the Blueprint for an AI Bill of Rights: Recommendations for Practitioners, Researchers, and Policy Makers
por: Oesterling, Alex, et al.
Publicado: (2024)

Which Models have Perceptually-Aligned Gradients? An Explanation via Off-Manifold Robustness
por: Srinivas, Suraj, et al.
Publicado: (2023)

Interpretability Needs a New Paradigm
por: Madsen, Andreas, et al.
Publicado: (2024)

Towards Unified Attribution in Explainable AI, Data-Centric AI, and Mechanistic Interpretability
por: Zhang, Shichang, et al.
Publicado: (2025)

Inference-Time Reward Hacking in Large Language Models
por: Khalaf, Hadi, et al.
Publicado: (2025)

Soft Best-of-n Sampling for Model Alignment
por: Verdun, Claudio Mayrink, et al.
Publicado: (2025)

Multi-Group Proportional Representation for Text-to-Image Models
por: Jung, Sangwon, et al.
Publicado: (2025)

Fair Machine Unlearning: Data Removal while Mitigating Disparities
por: Oesterling, Alex, et al.
Publicado: (2023)

Interpreting CLIP with Hierarchical Sparse Autoencoders
por: Zaigrajew, Vladimir, et al.
Publicado: (2025)

Characterizing Data Point Vulnerability via Average-Case Robustness
por: Han, Tessa, et al.
Publicado: (2023)

Interpretable and Steerable Concept Bottleneck Sparse Autoencoders
por: Kulkarni, Akshay, et al.
Publicado: (2025)

Hierarchical Concept Embedding & Pursuit for Interpretable Image Classification
por: Nguyen, Nghia, et al.
Publicado: (2026)

CASL: Concept-Aligned Sparse Latents for Interpreting Diffusion Models
por: He, Zhenghao, et al.
Publicado: (2026)

Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignment
por: Thasarathan, Harrish, et al.
Publicado: (2025)

Semantic Token Reweighting for Interpretable and Controllable Text Embeddings in CLIP
por: Kim, Eunji, et al.
Publicado: (2024)

Disentangling Latent Embeddings with Sparse Linear Concept Subspaces (SLiCS)
por: Li, Zhi, et al.
Publicado: (2025)

CLIP-UP: A Simple and Efficient Mixture-of-Experts CLIP Training Recipe with Sparse Upcycling
por: Wang, Xinze, et al.
Publicado: (2025)

Bi-ICE: An Inner Interpretable Framework for Image Classification via Bi-directional Interactions between Concept and Input Embeddings
por: Hong, Jinyung, et al.
Publicado: (2024)

Sparse CLIP: Co-Optimizing Interpretability and Performance in Contrastive Learning
por: Qin, Chuan, et al.
Publicado: (2026)

Sparse Autoencoders enable Robust and Interpretable Fine-tuning of CLIP models
por: Morelli, Fabian, et al.
Publicado: (2026)

Optimizing CLIP Models for Image Retrieval with Maintained Joint-Embedding Alignment
por: Schall, Konstantin, et al.
Publicado: (2024)

Towards Interpretable Soft Prompts
por: Patel, Oam, et al.
Publicado: (2025)

From Segments to Concepts: Interpretable Image Classification via Concept-Guided Segmentation
por: Eisenberg, Ran, et al.
Publicado: (2025)

CLIP-PAE: Projection-Augmentation Embedding to Extract Relevant Features for a Disentangled, Interpretable, and Controllable Text-Guided Face Manipulation
por: Zhou, Chenliang, et al.
Publicado: (2022)

Memorization In Stable Diffusion Is Unexpectedly Driven by CLIP Embeddings
por: Kim, Bumjun, et al.
Publicado: (2026)

From Weights to Concepts: Data-Free Interpretability of CLIP via Singular Vector Decomposition
por: Gentile, Francesco, et al.
Publicado: (2026)

Isotropic3D: Image-to-3D Generation Based on a Single CLIP Embedding
por: Liu, Pengkun, et al.
Publicado: (2024)

Interpreting CLIP: Insights on the Robustness to ImageNet Distribution Shifts
por: Crabbé, Jonathan, et al.
Publicado: (2023)

Enhancing Interpretability of Sparse Latent Representations with Class Information
por: Abiz, Farshad Sangari, et al.
Publicado: (2025)

Sparse Autoencoders for Interpretable Medical Image Representation Learning
por: Wesp, Philipp, et al.
Publicado: (2026)

Conceptualizing Embeddings: Sparse Disentanglement for Vision-Language Models
por: Kubaty, Piotr, et al.
Publicado: (2026)

DeCLIP: Decoding CLIP representations for deepfake localization
por: Smeu, Stefan, et al.
Publicado: (2024)

Concept-Centric Token Interpretation for Vector-Quantized Generative Models
por: Yang, Tianze, et al.
Publicado: (2025)

Interpretable Generative Models through Post-hoc Concept Bottlenecks
por: Kulkarni, Akshay, et al.
Publicado: (2025)

ConceptAttention: Diffusion Transformers Learn Highly Interpretable Features
por: Helbling, Alec, et al.
Publicado: (2025)