Related papers: Information Must Flow: Recursive Bootstrapping for…

Continual learning (CL) is a setting in which an agent has to learn from an incoming stream of data sequentially. CL performance evaluates the model's ability to continually learn and solve new problems with incremental available…

Contextual information is vital for accurate trajectory prediction. For instance, the intricate flying behavior of migratory birds hinges on their analysis of environmental cues such as wind direction and air pressure. However, the diverse…

Multi-agent collaboration holds great promise for enhancing the safety, reliability, and mobility of autonomous driving systems by enabling information sharing among multiple connected agents. However, existing multi-agent communication…

Recent studies on transformer-based language models show that they can answer questions by reasoning over knowledge provided as part of the context (i.e., in-context reasoning). However, since the available knowledge is often not filtered…

Interpreting and explaining the behavior of deep neural networks is critical for many tasks. Explainable AI provides a way to address this challenge, mostly by providing per-pixel relevance to the decision. Yet, interpreting such…

Catastrophic forgetting remains a central challenge in continual learning, where models are required to integrate new knowledge over time without losing what they have previously learned. In prior work, we introduced Cobweb/4V, a…

In-context learning (ICL) allows large models to adapt to tasks using a few examples, yet its extension to vision-language models (VLMs) remains fragile. Our analysis reveals that the fundamental limitation lies in an inductive gap, models…

Deploying reinforcement learning in safety critical domains, from autonomous vehicles to medical decision support, is constrained by failures arising when systems encounter unfamiliar conditions. We argue that the fundamental bottleneck is…

The transition from standard generative AI to \emph{reasoning-centric architectures}, exemplified by models capable of extensive Chain-of-Thought~(CoT) processing, marks a fundamental paradigm shift in system requirements. Unlike…

This paper investigates task-oriented communication for edge inference, where a low-end edge device transmits the extracted feature vector of a local data sample to a powerful edge server for processing. It is critical to encode the data…

Phase transitions abound in nature and society, and, from species extinction to stock market collapse, their prediction is of widespread importance. In earlier work we showed that Global Transfer Entropy, a general measure of information…

The Information Bottleneck method is a learning technique that seeks a right balance between accuracy and generalization capability through a suitable tradeoff between compression complexity, measured by minimum description length, and…

Traffic navigation services have gained widespread adoption in recent years. The route recommendations generated by these services often leads to severe congestion on urban streets, raising concerns from neighboring residents and city…

In this paper, we seek to measure how much information a component in a neural network could extract from the representations fed into it. Our work stands in contrast to prior probing work, most of which investigates how much information a…

Concept bottleneck models (CBMs) are interpretable models that first predict a set of semantically meaningful features, i.e., concepts, from observations that are subsequently used to condition a downstream task. However, the model's…

The core challenge in learning has shifted from knowledge acquisition to effective Self-Regulated Learning (SRL): planning, monitoring, and reflecting on one's learning. Existing digital tools, however, inadequately support metacognitive…

Large Language Models (LLMs) using Chain-of-Thought (CoT) prompting excel at complex reasoning but generate verbose thought processes with considerable redundancy, leading to increased inference costs and reduced efficiency. We introduce a…

In highly interactive driving scenes, trajectory prediction is conditioned on information from surrounding traffic participants such as cars and pedestrians. Our main contribution is a comprehensive analysis of state-of-the-art trajectory…

Modern deep neural networks have now reached human-level performance across a variety of tasks. However, unlike humans they lack the ability to explain their decisions by showing where and telling what concepts guided them. In this work, we…

Concept Bottleneck Models (CBMs) map the inputs onto a set of interpretable concepts (``the bottleneck'') and use the concepts to make predictions. A concept bottleneck enhances interpretability since it can be investigated to understand…