About
Research & Development
Improving Factual Accuracy in LLM Reasoning: The RELIANCE Framework
TLDR: RELIANCE is a new framework designed to improve the factual accuracy of Large Language Models' intermediate reasoning steps, addressing a critical vulnerability where...
Research & Development
Beyond Injustice: Understanding Arbitrariness as a Core System Function
TLDR: This research paper redefines arbitrariness not as a flaw or injustice, but as a fundamental functional mechanism that allows systems (linguistic, legal, social)...
Research & Development
SmartCourse: Enhancing Student Advising with Contextual AI
TLDR: SmartCourse is an AI-powered academic advising system for undergraduates, specifically Computer Science majors. It integrates student transcripts and degree plans with a local...
Research & Development
New Benchmark Reveals Spoken Dialogue Models Struggle with Ambiguity and Context in Real Conversations
TLDR: A new benchmark called C3 has been developed to evaluate Spoken Dialogue Models (SDMs) in complex, real-world conversations, focusing on challenges like ambiguity...
Research & Development
Unpacking the Limits of AI’s Multimodal Latent Spaces for Inverse Tasks
TLDR: A new study investigates whether AI models, designed for tasks like text-to-image generation, can effectively perform inverse tasks (e.g., inferring the original text...
Research & Development
ScanDiff: A Unified AI Model for Diverse Eye Movement Prediction
TLDR: ScanDiff is a new AI model that uses diffusion models to predict human eye movements (gaze scanpaths). Unlike older models that only predict...