AI Papers Podcast

AI Papers Podcast for 04/15/2024 COCONut: Modernizing COCO SegmentationScaling (Down) CLIP: A Comprehensive Analysis of Data, Architecture, and Training Strategies On the Robustness of Language Guidance for Low-Level Vision Tasks: Findings from Depth EstimationPre-training Small Base LMs with Fewer TokensProbing the 3D Awareness of Visual Foundation Models

AI Papers Podcast for 04/14/2024Direct Nash Optimization: Teaching Language Models to Self-Improve with  General Preferences: https://arxiv.org/abs/2404.03715No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency  Determines Multimodal Model Performance: https://arxiv.org/abs/2404.04125AutoWebGLM: Bootstrap And Reinforce A Large Language Model-based Web  Navigating Agent: https://arxiv.org/abs/2404.03648Stream of Search (SoS): Learning to Search in Language:

AI Papers Podcast for 04/13/2024OmniFusion Technical Report: https://arxiv.org/abs/2404.06212 LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders: https://arxiv.org/abs/2404.05961 InternLM-XComposer2-4KHD: A Pioneering Large Vision-Language Model Handling Resolutions from 336 Pixels to 4K HD: https://arxiv.org/abs/2404.06512 Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence: https://arxiv.org/abs/2404.05892 MiniCPM: Unveiling the Potential of Small

AI Papers Podcast 04/12/2024 RecurrentGemma: Moving Past Transformers for Efficient Open Language Models: https://arxiv.org/abs/2404.07839 WILBUR: Adaptive In-Context Learning for Robust and Accurate Web Agents: https://arxiv.org/abs/2404.05902 Ferret-v2: An Improved Baseline for Referring and Grounding with Large Language Models: https://arxiv.org/abs/2404.07973 Best Practices and Lessons Learned on Synthetic Data for Language Models: https://arxiv.org/abs/2404.07503 HGRN2: Gated

AI Papers Podcast for 04/12/2024 ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback: https://arxiv.org/abs/2404.07987 OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments: https://arxiv.org/abs/2404.07972 Rho-1: Not All Tokens Are What You Need: https://arxiv.org/abs/2404.07965 JetMoE: Reaching Llama2 Performance with 0.1M Dollars: https://arxiv.org/abs/2404.07413 Transferable and Principled Efficiency for Open-Vocabulary