118 citations · 253 across the 35 of their papers we have counts for
36 papers
Understanding Clustering in Slot Attention via Particle Dynamics
Vasudev Joy, Rajat Rasal, Avinash Kori +2
Studying attention through the lens of interacting particle dynamics has shown how token clustering can emerge from the underlying dynamics. We extend this perspective to slot atte…
A Unifying Perspective on Causal World Models: From Observations to Representations to Structure
Avinash Kori, Fabrizio Russo
World Models (WM) are increasingly seen as a foundation for intelligent agents that can predict, plan, and act beyond their training distribution. In this paper, we study WMs from…
Steering Optimisation Trajectories in Diffusion Representation Learning
Rajat Rasal, Avinash Kori, Tian Xia +1
We study why diffusion autoencoders can achieve similar image quality while learning substantially different latent structures. We trace this behaviour to optimisation dynamics; we…
Efficient bias mitigation in T2I diffusion models using Concept Graphs
Mansi, Avinash Kori, Francesco Leofante
Text-to-Image diffusion models often propagate harmful bias inherited from the training data. Existing bias mitigation techniques typically intervene only at the text encoder or pr…
Unified Multimodal Model for Brain MRI Imputation and Understanding
Zhiyun Song, Che Liu, Tian Xia +2
Multimodal large language models (MLLMs) hold great potential for medicine, as they inherit knowledge from LLM and allow multiple data modalities to be integrated, analysed and int…
Selective Fine-Tuning for Targeted and Robust Concept Unlearning
Mansi, Avinash Kori, Francesca Toni +1
Text guided diffusion models are used by millions of users, but can be easily exploited to produce harmful content. Concept unlearning methods aim at reducing the models' likelihoo…
Object-Centric Case-Based Reasoning via Argumentation
Gabriel de Olim Gaul, Adam Gould, Avinash Kori +1
We introduce Slot Attention Argumentation for Case-Based Reasoning (SAA-CBR), a novel neuro-symbolic pipeline for image classification that integrates object-centric learning via a…
Transparent Visual Reasoning via Object-Centric Agent Collaboration
Benjamin Teoh, Ben Glocker, Francesca Toni +1
A central challenge in explainable AI, particularly in the visual domain, is producing explanations grounded in human-understandable concepts. To tackle this, we introduce OCEAN (O…
Flow Stochastic Segmentation Networks
Fabio De Sousa Ribeiro, Omar Todd, Charles Jones +3
We introduce the Flow Stochastic Segmentation Network (Flow-SSN), a generative segmentation model family featuring discrete-time autoregressive and modern continuous-time flow vari…
Object-Centric Neuro-Argumentative Learning
Abdul Rahman Jacob, Avinash Kori, Emanuele De Angelis +3
Over the last decade, as we rely more on deep learning technologies to make critical decisions, concerns regarding their safety, reliability and interpretability have emerged. We i…
Diffusion Counterfactual Generation with Semantic Abduction
Rajat Rasal, Avinash Kori, Fabio De Sousa Ribeiro +2
Counterfactual image generation presents significant challenges, including preserving identity, maintaining perceptual quality, and ensuring faithfulness to an underlying causal mo…
Identifiable Object Representations under Spatial Ambiguities
Avinash Kori, Francesca Toni, Ben Glocker
Modular object-centric representations are essential for *human-like reasoning* but are challenging to obtain under spatial ambiguities, *e.g. due to occlusions and view ambiguitie…
Causal Representation Learning with Observational Grouping for CXR Classification
Rajat Rasal, Avinash Kori, Ben Glocker
Identifiable causal representation learning seeks to uncover the true causal relationships underlying a data generation process. In medical imaging, this presents opportunities to…
Factored Classifier-Free Guidance
Tian Xia, Fabio De Sousa Ribeiro, Rajat R Rasal +3
Counterfactual generation aims to simulate realistic hypothetical outcomes under causal interventions. Diffusion models have emerged as a powerful tool for this task, combining DDI…
Free Argumentative Exchanges for Explaining Image Classifiers
Avinash Kori, Antonio Rago, Francesca Toni
Deep learning models are powerful image classifiers but their opacity hinders their trustworthiness. Explanation methods for capturing the reasoning process within these classifier…
Explainable Reinforcement Learning for Formula One Race Strategy
Devin Thomas, Junqi Jiang, Avinash Kori +6
In Formula One, teams compete to develop their cars and achieve the highest possible finishing position in each race. During a race, however, teams are unable to alter the car, so…
Continuous Bayesian Model Selection for Multivariate Causal Discovery
Anish Dhir, Ruby Sedgwick, Avinash Kori +2
Current causal discovery approaches require restrictive model assumptions in the absence of interventional data to ensure structure identifiability. These assumptions often do not…
Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention
Avinash Kori, Francesco Locatello, Ainkaran Santhirasekaram +3
Learning modular object-centric representations is crucial for systematic generalization. Existing methods show promising object-binding capabilities empirically, but theoretical i…
A Causal Ordering Prior for Unsupervised Representation Learning
Avinash Kori, Pedro Sanchez, Konstantinos Vilouras +2
Unsupervised representation learning with variational inference relies heavily on independence assumptions over latent variables. Causal representation learning (CRL), however, arg…
Grounded Object Centric Learning
Avinash Kori, Francesco Locatello, Fabio De Sousa Ribeiro +2
The extraction of modular object-centric representations for downstream tasks is an emerging area of research. Learning grounded representations of objects that are guaranteed to b…
Explaining Image Classification with Visual Debates
Avinash Kori, Ben Glocker, Francesca Toni
An effective way to obtain different perspectives on any given topic is by conducting a debate, where participants argue for and against the topic. Here, we propose a novel debate…
Vector Quantisation for Robust Segmentation
Ainkaran Santhirasekaram, Avinash Kori, Mathias Winkler +2
The reliability of segmentation models in the medical domain depends on the model's robustness to perturbations in the input space. Robustness is a particular challenge in medical…
GLANCE: Global to Local Architecture-Neutral Concept-based Explanations
Avinash Kori, Ben Glocker, Francesca Toni
Most of the current explainability techniques focus on capturing the importance of features in input space. However, given the complexity of models and data-generating processes, t…
Hierarchical Symbolic Reasoning in Hyperbolic Space for Deep Discriminative Models
Ainkaran Santhirasekaram, Avinash Kori, Andrea Rockall +3
Explanations for \emph{black-box} models help us understand model decisions as well as provide information on model biases and inconsistencies. Most of the current explainability t…