Episode Details

Today's papers: - Does Local Video Understanding Transfer Across Encounters? The EgoGears Benchmark: https://arxiv.org/abs/2609.37938v1 - Social Choice Foundations for Simulation-Augmented Generation: https://arxiv.org/abs/2609.38287v1 - Probability is Not Enough: Exploring and Counting Divergent Tokens for Reasoning Uncertainty Quantification in LLMs: https://arxiv.org/abs/2609.38070v1 - A Benchmark & Dataset for Detecting AI-Manipulated Visual Evidence in the Court System: https://arxiv.org/abs/2609.37783v1 - AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation: https://arxiv.org/abs/2609.38142v1 This podcast is from Colin Davis (colin-davis.com) using Claude & Elevenlabs.