Cover Image

GCAV: A Global Concept Activation Vector Framework for Cross-Layer Consistency in Interpretability

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	He, Zhenghao, Sinha, Sanchit, Xiong, Guangzhi, Zhang, Aidong
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

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

Concept-RuleNet: Grounded Multi-Agent Neurosymbolic Reasoning in Vision Language Models
by: Sinha, Sanchit, et al.
Published: (2025)

ASCENT-ViT: Attention-based Scale-aware Concept Learning Framework for Enhanced Alignment in Vision Transformers
by: Sinha, Sanchit, et al.
Published: (2025)

COCO-Tree: Compositional Hierarchical Concept Trees for Enhanced Reasoning in Vision Language Models
by: Sinha, Sanchit, et al.
Published: (2025)

Retrieving Counterfactuals Improves Visual In-Context Learning
by: Xiong, Guangzhi, et al.
Published: (2026)

Attention-guided Fine-tuning of Multimodal Large Language Models Improves Chain-of-Thought Reasoning
by: Sinha, Sanchit, et al.
Published: (2026)

Rethinking Visual Attribution for Chest X-ray Reasoning in Large Vision Language Models
by: Xiong, Guangzhi, et al.
Published: (2026)

Chart-RVR: Reinforcement Learning with Verifiable Rewards for Explainable Chart Reasoning
by: Sinha, Sanchit, et al.
Published: (2025)

Reasoning Beyond Chain-of-Thought: A Latent Computational Mode in Large Language Models
by: He, Zhenghao, et al.
Published: (2026)

A Self-explaining Neural Architecture for Generalizable Concept Learning
by: Sinha, Sanchit, et al.
Published: (2024)

Structural Causality-based Generalizable Concept Discovery Models
by: Sinha, Sanchit, et al.
Published: (2024)

Toward Faithful Retrieval-Augmented Generation with Sparse Autoencoders
by: Xiong, Guangzhi, et al.
Published: (2025)

CoLiDR: Concept Learning using Aggregated Disentangled Representations
by: Sinha, Sanchit, et al.
Published: (2024)

ProtoNAM: Prototypical Neural Additive Models for Interpretable Deep Tabular Learning
by: Xiong, Guangzhi, et al.
Published: (2024)

Dictionary-based Framework for Interpretable and Consistent Object Parsing
by: Zhang, Tiezheng, et al.
Published: (2025)

LG-CAV: Train Any Concept Activation Vector with Language Guidance
by: Huang, Qihan, et al.
Published: (2024)

Causal Physics Steering in Video World Models via Concept Activation Vectors
by: Alam, Nahid
Published: (2026)

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

Exploring Interpretability for Visual Prompt Tuning with Cross-layer Concepts
by: Wang, Yubin, et al.
Published: (2025)

Can Cross-Layer Transcoders Replace Vision Transformer Activations? An Interpretable Perspective on Vision
by: Chatzoudis, Gerasimos, et al.
Published: (2026)

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

Towards Interpretable Geo-localization: a Concept-Aware Global Image-GPS Alignment Framework
by: Jia, Furong, et al.
Published: (2025)

Global Cross-Modal Geo-Localization: A Million-Scale Dataset and a Physical Consistency Learning Framework
by: Hu, Yutong, et al.
Published: (2026)

Neural Additive Experts: Context-Gated Experts for Controllable Model Additivity
by: Xiong, Guangzhi, et al.
Published: (2026)

LimeCross: Context-Conditioned Layered Image Editing with Structural Consistency
by: Morita, Ryugo, et al.
Published: (2026)

EgoMind: Activating Spatial Cognition through Linguistic Reasoning in MLLMs
by: Chen, Zhenghao, et al.
Published: (2026)

Optimizing Vision-Language Consistency via Cross-Layer Regional Attention Alignment
by: Wang, Yifan, et al.
Published: (2025)

Improving Network Interpretability via Explanation Consistency Evaluation
by: Wu, Hefeng, et al.
Published: (2024)

AdvST: Revisiting Data Augmentations for Single Domain Generalization
by: Zheng, Guangtao, et al.
Published: (2023)

Spuriousness-Aware Meta-Learning for Learning Robust Classifiers
by: Zheng, Guangtao, et al.
Published: (2024)

4DGS-Craft: Consistent and Interactive 4D Gaussian Splatting Editing
by: Liu, Lei, et al.
Published: (2025)

SPARC: Concept-Aligned Sparse Autoencoders for Cross-Model and Cross-Modal Interpretability
by: Nasiri-Sarvi, Ali, et al.
Published: (2025)

DCG-Net: Dual Cross-Attention with Concept-Value Graph Reasoning for Interpretable Medical Diagnosis
by: Dagnaw, Getamesay, et al.
Published: (2026)

Robust Concept Erasure Using Task Vectors
by: Pham, Minh, et al.
Published: (2024)

MedConcept: Unsupervised Concept Discovery for Interpretability in Medical VLMs
by: Haque, Md Rakibul, et al.
Published: (2026)

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

ProtoSolo: Interpretable Image Classification via Single-Prototype Activation
by: Peng, Yitao, et al.
Published: (2025)

Enhancing Cross-View Geo-Localization Generalization via Global-Local Consistency and Geometric Equivariance
by: Wang, Xiaowei, et al.
Published: (2025)

Where Concept Erasure Should Occur: Concept-Layer Alignment in Text-to-Video Diffusion Models
by: Xie, Yiwei, et al.
Published: (2026)

WWW: A Unified Framework for Explaining What, Where and Why of Neural Networks by Interpretation of Neuron Concepts
by: Ahn, Yong Hyun, et al.
Published: (2024)