Despite their many successes, transformer-based large language models (LLMs) continue to struggle with tasks that require complex reasoning over large parts of their input. We argue that these failures arise due to capacity limits on the accurate flow of information within LLMs. To formalize this…
Generative modeling, representation learning, and classification are three core problems in machine learning (ML), yet their state-of-the-art (SoTA) solutions remain largely disjoint. In this paper, we ask: Can a unified principle address all three? Such unification could simplify ML pipelines and foster greater synergy across…
Organisms constantly pivot between tasks such as evading predators, foraging, traversing rugged terrain, and socializing, often within milliseconds. Remarkably, they preserve knowledge of once-learned environments sans catastrophic forgetting, a phenomenon neuroscientists hypothesize, is due to a singular neural circuitry dynamically overlayed by neuromodulatory agents such…
Progress in computer use agents (CUAs) has been constrained by the absence of large and high-quality datasets that capture how humans interact with a computer. While LLMs have thrived on abundant textual data, no comparable corpus exists for CUA trajectories. To address these gaps, we…
AI-assisted report generation offers the opportunity to reduce radiologists'workload stemming from expanded screening guidelines, complex cases and workforce shortages, while maintaining diagnostic accuracy. In addition to describing pathological findings in chest X-ray reports, interpreting lines and tubes (L&T) is demanding and repetitive for radiologists, especially…
Serving large generative models such as LLMs and multi-modal transformers requires balancing user-facing SLOs (e.g., time-to-first-token, time-between-tokens) with provider goals of efficiency and cost reduction. Existing solutions rely on static provisioning or model-level autoscaling, both of which treat the model as a monolith. This coarse-grained…
Empathy is a critical factor in fostering positive user experiences in conversational AI. While models can display empathy, it is often generic rather than tailored to specific tasks and contexts. In this work, we introduce a novel framework for developing and evaluating context-specific empathetic large…
Developing end-to-end bioinformatics workflows is challenging, demanding deep expertise in both genomics and computational techniques. While large language models (LLMs) provide some assistance, they often lack the nuanced guidance required for complex bioinformatics tasks, and are resource-intensive. We thus propose a multi-agent system built on…
While Large Language Models (LLMs) are increasingly used in agentic frameworks to assist individual users, there is a growing need for agents that can proactively manage complex, multi-party collaboration. Systematic evaluation methods for such proactive agents remain scarce, limiting progress in developing AI that can…
Artificial intelligence (AI) holds tremendous potential but also poses consequential risks. Regulation frameworks like the EU AI Act aim to mitigate these risks, yet organizations struggle to understand and operationalize Responsible AI (RAI). We introduce the RAI Organizational Maturity (RAI-OM) framework as an initial step…
Advances in artificial intelligence (AI)–assisted protein engineering are enabling breakthroughs in the life sciences but also introduce new biosecurity challenges. Synthesis of nucleic acids is a choke point in AI-assisted protein engineering pipelines. Thus, an important focus for efforts to enhance biosecurity given AI-enabled capabilities…
As large language models (LLMs) are increasingly used for code generation, concerns over the security risks have grown substantially. Early research has primarily focused on red teaming, which aims to uncover and evaluate vulnerabilities and risks of CodeGen models. However, progress on the blue teaming…
