12 papers
Visual Semantic Entropy: Do Vision Language Models Recognize Visual Ambiguity?
Ta Duc Huy, Trang Nguyen, Townim Chowdhury +5
Vision-language models can produce confident answers on visually ambiguous inputs, resulting in biased predictions. Common entropy-based methods, such as Semantic Entropy (SE), rel…
AR&D: A Framework for Retrieving and Describing Concepts for Interpreting AudioLLMs
Townim Faisal Chowdhury, Ta Duc Huy, Siqi Pan +2
Despite strong performance in audio perception tasks, large audio-language models (AudioLLMs) remain opaque to interpretation. A major factor behind this lack of interpretability i…
From Healthy Scans to Annotated Tumors: A Tumor Fabrication Framework for 3D Brain MRI Synthesis
Nayu Dong, Townim Chowdhury, Hieu Phan +3
The scarcity of annotated Magnetic Resonance Imaging (MRI) tumor data presents a major obstacle to accurate and automated tumor segmentation. While existing data synthesis methods…
Looking in the mirror: A faithful counterfactual explanation method for interpreting deep image classification models
Townim Faisal Chowdhury, Vu Minh Hieu Phan, Kewen Liao +5
Counterfactual explanations (CFE) for deep image classifiers aim to reveal how minimal input changes lead to different model decisions, providing critical insights for model interp…
AdaCBM: An Adaptive Concept Bottleneck Model for Explainable and Accurate Diagnosis
Townim F. Chowdhury, Vu Minh Hieu Phan, Kewen Liao +5
The integration of vision-language models such as CLIP and Concept Bottleneck Models (CBMs) offers a promising approach to explaining deep neural network (DNN) decisions using conc…
CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation
Townim Faisal Chowdhury, Kewen Liao, Vu Minh Hieu Phan +7
Deep Neural Networks (DNNs) are widely used for visual classification tasks, but their complex computation process and black-box nature hinder decision transparency and interpretab…
ChatGPT-guided Semantics for Zero-shot Learning
Fahimul Hoque Shubho, Townim Faisal Chowdhury, Ali Cheraghian +3
Zero-shot learning (ZSL) aims to classify objects that are not observed or seen during training. It relies on class semantic description to transfer knowledge from the seen classes…
Prompt-guided Scene Generation for 3D Zero-Shot Learning
Majid Nasiri, Ali Cheraghian, Townim Faisal Chowdhury +3
Zero-shot learning on 3D point cloud data is a related underexplored problem compared to its 2D image counterpart. 3D data brings new challenges for ZSL due to the unavailability o…
Rethinking Task-Incremental Learning Baselines
Md Sazzad Hossain, Pritom Saha, Townim Faisal Chowdhury +3
It is common to have continuous streams of new data that need to be introduced in the system in real-world applications. The model needs to learn newly added capabilities (future t…
Few-shot Class-incremental Learning for 3D Point Cloud Objects
Townim Chowdhury, Ali Cheraghian, Sameera Ramasinghe +3
Few-shot class-incremental learning (FSCIL) aims to incrementally fine-tune a model (trained on base classes) for a novel set of classes using a few examples without forgetting the…
Learning without Forgetting for 3D Point Cloud Objects
Townim Chowdhury, Mahira Jalisha, Ali Cheraghian +1
When we fine-tune a well-trained deep learning model for a new set of classes, the network learns new concepts but gradually forgets the knowledge of old training. In some real-lif…
Zero-Shot Learning on 3D Point Cloud Objects and Beyond
Ali Cheraghian, Shafinn Rahman, Townim F. Chowdhury +2
Zero-shot learning, the task of learning to recognize new classes not seen during training, has received considerable attention in the case of 2D image classification. However, des…