12 accepted papers
Most multimodal learning assumes redundant views (such as image–caption pairs), yet many applications require combining complementary modalities that provide distinct evidence (such as an X-ray and medical history). We term this setting *Complementary Multimodal Classification* (CMC). In CMC, existi…
Memory and computation remain core bottlenecks in long-horizon LLM inference due to the quadratic cost of self-attention and the ever-growing key-value (KV) cache. Existing strategies for memory-bounded inference, such as quantization, offloading, or heuristic KV eviction, either incur high orchestr…
As digital platforms redefine educational paradigms, ensuring interactivity remains vital for effective learning. This paper explores using Multimodal Large Language Models (MLLMs) to automatically respond to student questions from online lectures - a novel question answering task of real world sign
Data distillation and coresets have emerged as popular approaches to generate a smaller representative set of samples for downstream learning tasks to handle large-scale datasets. At the same time, machine learning is being increasingly applied to decision-making processes at a societal level, makin…
The Rashomon effect is a mixed blessing in responsible machine learning. It enhances the prospects of finding models that perform well in accuracy while adhering to ethical standards, such as fairness or interpretability. Conversely, it poses a risk to the credibility of machine decisions through pr…
With the growing use of machine learning (ML) models in critical domains such as finance and healthcare, the need to offer recourse for those adversely affected by the decisions of ML models has become more important; individuals ought to be provided with recommendations on actions to take for impro…
Three key properties that are desired of trustworthy machine learning models deployed in high-stakes environments are fairness, explainability, and an ability to account for various kinds of "drift". While drifts in model accuracy have been widely investigated, drifts in fairness metrics over time r…
Learning effective reinforcement learning (RL) policies to solve real-world complex tasks can be quite challenging without a high-fidelity simulation environment. In most cases, we are only given imperfect simulators with simplified dynamics, which inevitably lead to severe sim-to-real gaps in RL po…
Given a propositional formula ψ, the model counting problem, also referred to as #SAT, seeks to compute the number of satisfying assignments (or models) of ψ. Modern search-based model counting algorithms are built on conflict-driven clause learning, combined with the caching of certain subformulas…
As more companies and governments build and use machine learning models to automate decisions, there is an ever-growing need to monitor and evaluate these models' behavior once they are deployed. Our team at CognitiveScale has developed a toolkit called Cortex Certifai to answer this need. Cortex…
The development of compressed sensing (CS) techniques for magnetic resonance imaging (MRI) is enabling a speedup of MRI scanning. To increase the incoherence in the sampling, a random selection of points on the k-space is deployed and a continuous trajectory is obtained by solving a traveling salesm…
A novel class of k-space trajectories for magnetic resonance imaging (MRI) sampling using space filling curves (SFCs) is presented here. More specifically, Peano, Hilbert and Sierpinski curves are used. We propose 1-shot and 4-shot variable density SFCs by utilizing the space coverage provided by SF…