Daily Paper Cast
Jingwen Liang, Gengyu Wang
We update every weekday to discuss highest-voted papers from Huggingface Daily Paper (https://huggingface.co/papers). Both the podcast scripts and audio are generated by AI. Feedback and suggestions are welcome! Email us: [email protected]:Jingwen Liang, 3D ML, https://www.linkedin.com/in/jingwen-liang/Gengyu Wang, LLM ML, http://wanggengyu.comListen on: Spotify: https://open.spotify.com/show/21nrhmdaA8qoBiH8q03NXLApple Podcast: https://podcasts.apple.com/us/podcast/daily-paper-...
Episodes (1777)
Self-Forcing++: Towards Minute-Scale High-Quality Video Generation
Daily Paper Cast
ExGRPO: Learning to Reason from Experience
Daily Paper Cast
StealthAttack: Robust 3D Gaussian Splatting Poisoning via Density-Guided Illusio...
Daily Paper Cast
Interactive Training: Feedback-Driven Neural Network Optimization
Daily Paper Cast
ModernVBERT: Towards Smaller Visual Document Retrievers
Daily Paper Cast
StockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets?
Daily Paper Cast
DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Re...
Daily Paper Cast
GEM: A Gym for Agentic LLMs
Daily Paper Cast
VLA-RFT: Vision-Language-Action Reinforcement Fine-tuning with Verified Rewards...
Daily Paper Cast
Knapsack RL: Unlocking Exploration of LLMs via Optimizing Budget Allocation
Daily Paper Cast
PIPer: On-Device Environment Setup via Online Reinforcement Learning
Daily Paper Cast
SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM We...
Daily Paper Cast
ACON: Optimizing Context Compression for Long-horizon LLM Agents
Daily Paper Cast
MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use
Daily Paper Cast
The Dragon Hatchling: The Missing Link between the Transformer and Models of the...
Daily Paper Cast
Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified Self-Play
Daily Paper Cast
Winning the Pruning Gamble: A Unified Approach to Joint Sample and Token Pruning...
Daily Paper Cast
TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
Daily Paper Cast
Learning to See Before Seeing: Demystifying LLM Visual Priors from Language Pre-...
Daily Paper Cast
OceanGym: A Benchmark Environment for Underwater Embodied Agents
Daily Paper Cast
More Thought, Less Accuracy? On the Dual Nature of Reasoning in Vision-Language...
Daily Paper Cast
Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effect...
Daily Paper Cast
DC-VideoGen: Efficient Video Generation with Deep Compression Video Autoencoder
Daily Paper Cast
SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear At...
Daily Paper Cast
StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs
Daily Paper Cast
Multiplayer Nash Preference Optimization
Daily Paper Cast
RealUnify: Do Unified Models Truly Benefit from Unification? A Comprehensive Ben...
Daily Paper Cast
Beyond the Exploration-Exploitation Trade-off: A Hidden State Approach for LLM R...
Daily Paper Cast
OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Edit...
Daily Paper Cast
SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer
Daily Paper Cast
Democratizing AI scientists using ToolUniverse
Daily Paper Cast
Visual Jigsaw Post-Training Improves MLLMs
Daily Paper Cast
When Does Reasoning Matter? A Controlled Study of Reasoning's Contribution to Mo...
Daily Paper Cast
LongLive: Real-time Interactive Long Video Generation
Daily Paper Cast
Quantile Advantage Estimation for Entropy-Safe Reasoning
Daily Paper Cast
EPO: Entropy-regularized Policy Optimization for LLM Agents Reinforcement Learni...
Daily Paper Cast
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Docum...
Daily Paper Cast
ReviewScore: Misinformed Peer Review Detection with Large Language Models
Daily Paper Cast
Variational Reasoning for Language Models
Daily Paper Cast
Language Models Can Learn from Verbal Feedback Without Scalar Rewards
Daily Paper Cast
MesaTask: Towards Task-Driven Tabletop Scene Generation via 3D Spatial Reasoning
Daily Paper Cast
CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning
Daily Paper Cast
No Prompt Left Behind: Exploiting Zero-Variance Prompts in LLM Reinforcement Lea...
Daily Paper Cast
VCRL: Variance-based Curriculum Reinforcement Learning for Large Language Models
Daily Paper Cast
SciReasoner: Laying the Scientific Reasoning Ground Across Disciplines
Daily Paper Cast
MMR1: Enhancing Multimodal Reasoning with Variance-Aware Sampling and Open Resou...
Daily Paper Cast
Tree Search for LLM Agent Reinforcement Learning
Daily Paper Cast
Seedream 4.0: Toward Next-generation Multimodal Image Generation
Daily Paper Cast
Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets
Daily Paper Cast
AutoIntent: AutoML for Text Classification
Daily Paper Cast