About
Research & Development
Unifying the Measurement of Proactive AI Dialogue
TLDR: ProactiveEval is a new unified framework for evaluating large language models' proactive dialogue capabilities. It addresses fragmented evaluations by decomposing proactive dialogue into...
Research & Development
New Benchmark Reveals Challenges for AI in Understanding Fine Acoustic Details
TLDR: The WoW-Bench research introduces a new benchmark to evaluate how well large audio-language models (LALMs) perceive and process fine-grained acoustic details, particularly using...
Research & Development
ExpertSim: Accelerating Particle Detector Simulations at CERN with Generative AI
TLDR: ExpertSim is a novel deep learning method that utilizes a Mixture-of-Generative-Experts architecture to significantly enhance the speed and accuracy of particle detector simulations,...
Research & Development
Unraveling Video Actions: A Structured Approach to Explainable Causal AI
TLDR: The "ChainReaction!" research introduces a new method for Causal-Why Video Question Answering (VideoQA) that makes AI reasoning more transparent and accurate. Instead of...
Research & Development
Generative AI and Human Cognition: Risks, Opportunities, and the Future of Thought
TLDR: This article synthesizes findings from the CHI 2025 Tools for Thought Workshop, exploring how Generative AI (GenAI) impacts human cognition. It details both...
Research & Development
AI and Logic Unite for Trustworthy Tax Assistance
TLDR: This research paper introduces a neuro-symbolic AI approach that integrates large language models (LLMs) with a symbolic solver to provide accurate and auditable...