Publications

Why Far Looks Up: Probing Spatial Representation in Vision-Language Models

Cheolhong Min, Jaeyun Jung, Daeun Lee, Hyeonseong Jeon, Yu Su, Jonathan Tremblay, Chan Hee Song, Jaesik Park
Paper

VisLens: Single-Pass Interpretable Visual Search for Multimodal LLMs

Jingyi He, Sanghwan Kim, Zeynep Akata

Vision-Language Models Trust What They Can Discern: Image and Representational Properties Governing Modality Arbitration

Vijay Vankadaru, Vasha DuTell

SIC: Similarity-Based Interpretable Image Classification with Neural Networks

Tom Nuno Wolf, Emre Kavak, Fabian Bongratz, Christian Wachinger
Paper

Seeing Before Answering: Training-Free Visual Layer Profiling for Vision-Language Models

Ruchen Liu, Yi Yang, Yiming Xu, Michael Ying Yang, Monika Sester, Bodo Rosenhahn

Queryable Is Not Grounded: Reference-Anchored Faithfulness Evaluation for Foundation Vision-Language Models

Anna Pramod

ProtoMappingNet: Interpretable Hierarchical Prototypes through Relational Prototype Mappings

Jaehun Park, Jongmin Lim, Soobin Cha, Kwangsu Kim
Paper

Position: We need theory-grounded explainability for abstract tasks to quantify model bias.

Katharina Prasse, Margret Keuper

On the Faithfulness of Post-Hoc Concept Bottleneck Models

Laines Schmalwasser, Jan Blunk, Niklas Penzel, Julia Niebling, Joachim Denzler
Paper

Modeling Visual Flow Leakage to Alleviate Hallucination in Large Vision-Language Models

Qi Bi, Jingjun Yi, Remco Veltkamp, Albert Ali Salah

Logit Lens Supervision for Patch-Level Explanations in Vision-Language Models

Parsa Esmaeilkhani, Longin Jan Latecki

LogicCBMs: Logic-Enhanced Concept-Based Learning

Deepika SN Vemuri, Gautham Bellamkonda, Aditya Pola, Vineeth N Balasubramanian
Paper

Learn to Rank: Visual Attribution by Learning Importance Ranking

David Schinagl, Christian Fruhwirth-Reisinger, Alexander Prutsch, Samuel Schulter, Horst Possegger
Paper

Influence Functions for Flow-Matching Generative Models: A K-FAC Port and an Honest Scale Study

Charchit Sharma

From Interpretability Methods to Interpretable Models

Julien Colin, Nuria M Oliver, Thomas Serre

From Image Latent Space to Fuzzy Rules: Interpretable Analysis of Gastrointestinal Foundation Model

Michael Vasilakakis, Dimitris Iakovidis

From Drop-off to Recovery: A Mechanistic Analysis of Segmentation in MLLMs

Boyong Wu, Sanghwan Kim, Zeynep Akata
Paper

Formal Concept Lattices are Good Semantic Scaffolds for Concept-Based Learning

Deepika SN Vemuri, Sayanta Adhikari, Ankit Saha, Krishn Vishwas Kher, Vineeth N Balasubramanian
Paper

Faithful Is Not Trustworthy: The Region of Interest Is Underdetermined in MIL Attribution

Robin Birker, Katarzyna Bozek

EXPOSE: Explainable and Domain-Robust Embeddings from Pathology Vision Foundation Models using Sparse Autoencoders

Anja Witte, Maximilian Lennartz, Jan Baumbach, Guido Sauter, Stefan Bonn, Patrick Fuhlert, Marina Zimmermann

EXPL-FR: Explaining Face Recognition Models via Vision-Language Alignment

Guray Ozgur, Mustafa Efe Tamyapar, Naser Damer, Fadi Boutros

Evaluating the Interpretability of Sparse Autoencoders with Concept Annotations

Jonas Klotz, Cassio F. Dantas, Pallavi Jain, Diego Marcos, Begüm Demir
Paper

Emergent Object Binding Has a Finite Spatial Horizon

Mayank Singal

DoCoG: Mask-based Multi-Type Grounded Chain-of-Thought for Document QA

Sai Madhusudan Gunda, Jyothi Swaroopa Jinka, Hrithik Sagar Rachakonda, Aryan Jain, Venkata Kesav Venna, Anirudh Srinivasan, Ravi Kiran Sarvadevabhatla
Paper

DiMaS: Distribution Matching for Steering Vision-Language-Action Models

Pegah Khayatan, Sara Meziane, Jayneel Parekh, Matthieu Cord

CFM: Language-aligned Concept Foundation Model for Vision

Kai Wittenmayer, Sukrut Rao, Amin Parchami-Araghi, Bernt Schiele, Jonas Fischer
Paper

CBX-Bench: A Human-Aligned MLLM Council for Benchmarking Concept Bottleneck Model Explanations

Yusuf Meric Karadag, Gulay Oklan, Seref Baris Cagliyan, Umut Ozdemir, Emre Akbas

Attention (as Discrete-Time Markov) Chains

Yotam Erel, Olaf Dünkel, Rishabh Dabral, Vladislav Golyanik, Christian Theobalt, Amit H. Bermano
Paper

Ask Twice, Look Twice: Prompt Echoing Resolves the Question-First Paradox in Vision-Language Models

Rakshanda Hassan Abhinandan, John Galeotti, Deva Ramanan, Gautam Rajendrakumar Gare