Show Notes
Today's papers: - Nemotron 3.5 Super and Lightning: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Models for Agentic AI: https://arxiv.org/abs/2610.38095v1 - PerceptRL: Making Visual Perception Rewardable in MLLMs from Answers to Interactions: https://arxiv.org/abs/2610.37471v1 - WorldFlow: Joint Video and 3D Motion Prediction via Large-Scale Pretraining: https://arxiv.org/abs/2610.39278v1 - Assessing Language Models for Scientific Applications: Inertial Confinement Fusion as a Use Case: https://arxiv.org/abs/2610.38024v1 - Zooming without Zooming v2: From Seeing Details to Connecting Evidence for Generalizable Multimodal Perception: https://arxiv.org/abs/2610.37037v1
This podcast is from Colin Davis (colin-davis.com) using Claude & Elevenlabs.
More from DailyArxiv - AI Research Podcast
Today's papers: - CoS-VQ: Coordinated Shared-Scale Vector Quantization for Large Language Models: https://arxiv.org/abs/2610.37336v1 - RoboAuto: Autonomous Robot Program Search through Multi-Level Physical Understanding: https://arxiv.org/abs/2610.36533v1 - Your Agent Acts Differently When You're Around: How User Presence Changes Agent Behavior in Negotiation: https://arxiv.org/abs/2610.36435v1 - ATEMPO: Amortized Transfer Entropy Model Per-lag Over-Time: https://arxiv.org/abs/2610.36536v1 - Concordat: Typed Sessions for LLM Agents: https://arxiv.org/abs/2610.37461v1
Today's papers: - Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol: https://arxiv.org/abs/2610.36108v1 - Patient-specific rubric-based evaluation of large language models across longitudinal cancer care: https://arxiv.org/abs/2610.35847v1 - Timestral: https://arxiv.org/abs/2610.37585v1 - Pretraining Knows What, Simulation Teaches How: Compositional Generalization in a Sim-Trained Robot Agent: https://arxiv.org/abs/2610.35817v2 - VideoLVR: Latent Visual Reasoning over Grounded Video Evidence: https://arxiv.org/abs/2610.37613v1
Today's papers: - InferSlice: A Reference Architecture for SLO-driven LLM Inference over a Telco Wide-Area Network: https://arxiv.org/abs/2610.32638v1 - Semantic Representation for Agentic Self-Play in Automated Laboratories: https://arxiv.org/abs/2610.32755v1 - The LLENS Platform for On-Premises Large Language Model Agents in a Japanese University Hospital: Development and Evaluation Study: https://arxiv.org/abs/2610.32151v2 - Governing Enterprise AI Traffic at Scale: Design and Operation of a Policy-Driven, Multi-Vendor LLM Gateway on Envoy: https://arxiv.org/abs/2610.35509v1 - FlashScript: Compiling Recurring Agent Workflows into Reusable Neuro-Symbolic Programs: https://arxiv.org/abs/2610.33876v1
Today's papers: - CoWriter: Guiding Methodological and Synergetic Human-LLM Co-Authoring: https://arxiv.org/abs/2610.31277v1 - A Mathematical Theory of the Jacobian Lens: https://arxiv.org/abs/2610.31955v1 - Anticipating Social Consequences with Distributional World Models: https://arxiv.org/abs/2610.30657v1 - Directing to Diversify: Multi-Directional Search for Automatic Heuristic Design: https://arxiv.org/abs/2610.31048v1 - The Helpful Agent Problem: When Emergent Cooperation Among LLM Agents Bypasses Intended Workflows: https://arxiv.org/abs/2610.31985v1
Today's papers: - Parallel Thinking for Large Language Model with Path Diversity: https://arxiv.org/abs/2610.29501v1 - Enhancing Cross-Grade Generalization in Banana Variety Classification via Hybrid CNN-Transformers and Knowledge Self-Distillation: https://arxiv.org/abs/2610.29008v1 - Same Beliefs, Different Policies: Learning Reusable Representations for Early Stopping in LLM Reasoning: https://arxiv.org/abs/2610.30450v1 - Mixing Matters: Evidence Position Bias Across Sequence Mixers in Long-Context Question Answering: https://arxiv.org/abs/2610.31135v1 - Multimodal Embodied Agents: A Unified Capability-centric Perspective from Computer-Use to Robot-Use: https://arxiv.org/abs/2610.30108v1