Shows / Unzip
Unzip
Your guide to the latest AI and machine learning research. We unpack complex papers into actionable insights for practitioners and enthusiasts alike.
Episodes
UnzipPartition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and World Models
## Episode Summary In this episode, we cover: - **Partition the Support, Reconstruct the Residual: Training-Free Sparse Attention for Video Generation and…
Unzip4DAnyone: Create Anyone in 4D from a Casual Monocular Video
## Episode Summary In this episode, we cover: - **4DAnyone: Create Anyone in 4D from a Casual Monocular Video** (arXiv) - [Read…
UnzipNARU: A Benchmark for NARrative Evolution and Cultural Nuance Understanding in Japanese Extreme Long Video
## Episode Summary In this episode, we cover: - **NARU: A Benchmark for NARrative Evolution and Cultural Nuance Understanding in Japanese Extreme Long Video**…
UnzipThinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See
## Episode Summary In this episode, we cover: - **Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See** (Hugging Face…
UnzipZetta ζ: An Efficient Closed-Loop Embodied Harness for Self-Evolving Physical Intelligence
## Episode Summary In this episode, we cover: - **Zetta ζ: An Efficient Closed-Loop Embodied Harness for Self-Evolving Physical Intelligence** (Hugging Face…
UnzipHarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety
## Episode Summary In this episode, we cover: - **HarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety** (Hugging Face Daily) - [Read…
UnzipDon't Drop the BATON: Long-Horizon Robot Manipulation via Agentic Subtask Exploration and Transition-aware Memory
## Episode Summary In this episode, we cover: - **Don't Drop the BATON: Long-Horizon Robot Manipulation via Agentic Subtask Exploration and Transition-aware…
UnzipDFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible Post-Training Data
## Episode Summary In this episode, we cover: - **DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible…
UnzipThe data geometry of masking diffusion: Certified-optimal schedules via unmasking growth complexity
## Episode Summary In this episode, we cover: - **The data geometry of masking diffusion: Certified-optimal schedules via unmasking growth complexity** (arXiv)…
UnzipSKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models
## Episode Summary In this episode, we cover: - **SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models**…
UnzipMitigating Gender Bias in English to Romanian Machine Translation
## Episode Summary In this episode, we cover: - **Mitigating Gender Bias in English to Romanian Machine Translation** (Hugging Face Daily) - [Read…
UnzipMechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence
## Episode Summary In this episode, we cover: - **Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence** (Hugging Face…
UnzipCan LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives
## Episode Summary In this episode, we cover: - **Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives**…
UnzipContextMaster: Interactive Multi-Shot Video Creation via Fixed-Budget Sparse Context Routing
## Episode Summary In this episode, we cover: - **ContextMaster: Interactive Multi-Shot Video Creation via Fixed-Budget Sparse Context Routing** (Hugging Face…
UnzipInterpretable MEG Decoding of Perceived Speech: Cortical Sources and the Stimulus Features That Drive Retrieval
## Episode Summary In this episode, we cover: - **Interpretable MEG Decoding of Perceived Speech: Cortical Sources and the Stimulus Features That Drive…
UnzipDataSpace: Benchmarking Data Agents for Verifiable Analytics over Heterogeneous Workspaces
## Episode Summary In this episode, we cover: - **DataSpace: Benchmarking Data Agents for Verifiable Analytics over Heterogeneous Workspaces** (Hugging Face…
UnzipAVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities
## Episode Summary In this episode, we cover: - **AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities** (Hugging Face Daily) - [Read…
UnzipWorldCup Arena: Prospective, Leakage-Free Evaluation of Frontier LLMs on a Live Tournament
## Episode Summary In this episode, we cover: - **WorldCup Arena: Prospective, Leakage-Free Evaluation of Frontier LLMs on a Live Tournament** (arXiv) - [Read…
UnzipModel or Harness? An Interaction-Centric Taxonomy for Localizing Agent Failures
## Episode Summary In this episode, we cover: - **Model or Harness? An Interaction-Centric Taxonomy for Localizing Agent Failures** (Hugging Face Daily) -…
UnzipSAF-OPD: Stable Advantage Fusion for On-Policy Distillation
## Episode Summary In this episode, we cover: - **SAF-OPD: Stable Advantage Fusion for On-Policy Distillation** (Hugging Face Daily) - [Read…
UnzipFairness Pruning: Locating Demographic Bias in GLU-MLP Layers via Differential Activations
## Episode Summary In this episode, we cover: - **Fairness Pruning: Locating Demographic Bias in GLU-MLP Layers via Differential Activations** (Hugging Face…
UnzipMulti-Head Attention Residuals
## Episode Summary In this episode, we cover: - **Multi-Head Attention Residuals** (Hugging Face Daily) - [Read more](https://huggingface.co/papers/2607.27230)…
UnzipOSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models
## Episode Summary In this episode, we cover: - **OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models** (arXiv) - [Read…
UnzipSpecFirst: Behavioral Specification Elicitation as a First-Class Step in Agent-Based Program Synthesis from Scratch
## Episode Summary In this episode, we cover: - **SpecFirst: Behavioral Specification Elicitation as a First-Class Step in Agent-Based Program Synthesis from…
UnzipMODUS: Decoder-Only Any-to-Any Modeling of Diverse Modalities
## Episode Summary In this episode, we cover: - **MODUS: Decoder-Only Any-to-Any Modeling of Diverse Modalities** (Hugging Face Daily) - [Read…
UnzipTILT: Improving Compositional Generation in Diffusion Models with a Model-Intrinsic Reward
## Episode Summary In this episode, we cover: - **TILT: Improving Compositional Generation in Diffusion Models with a Model-Intrinsic Reward** (Hugging Face…
UnzipScaling Native Multimodal Pre-Training From Scratch
## Episode Summary In this episode, we cover: - **Scaling Native Multimodal Pre-Training From Scratch** (Hugging Face Daily) - [Read…
UnzipFinanceComplexQA: Benchmarking Agentic Reasoning on Industrial-grade Financial Documents
## Episode Summary In this episode, we cover: - **FinanceComplexQA: Benchmarking Agentic Reasoning on Industrial-grade Financial Documents** (Hugging Face…
UnzipColor Pass-Through via Camera-Display Coupling
## Episode Summary In this episode, we cover: - **Color Pass-Through via Camera-Display Coupling** (Hugging Face Daily) - [Read…
UnzipStreaming Multi-Agent Autoregressive Diffusion Model with World State Registers
## Episode Summary In this episode, we cover: - **Streaming Multi-Agent Autoregressive Diffusion Model with World State Registers** (arXiv) - [Read…
UnzipDocOps: A Verifiable Benchmark for Autonomous Agents in Complex Document Operations
## Episode Summary In this episode, we cover: - **DocOps: A Verifiable Benchmark for Autonomous Agents in Complex Document Operations** (Hugging Face Daily) -…
UnzipAgents in the Wild: Where Research Meets Deployment
## Episode Summary In this episode, we cover: - **Agents in the Wild: Where Research Meets Deployment** (arXiv) - [Read…
UnzipGigaChat Audio: Time-aware Large Audio Language Model
## Episode Summary In this episode, we cover: - **GigaChat Audio: Time-aware Large Audio Language Model** (Hugging Face Daily) - [Read…
UnzipAn Exam for Active Observers
## Episode Summary In this episode, we cover: - **An Exam for Active Observers** (arXiv) - [Read more](http://arxiv.org/abs/2607.16165v1) - **When Does Muon…
UnzipAsySplat: Efficient Asymmetric 3D Gaussian Splatting for Long-Sequence Scene Modeling
## Episode Summary In this episode, we cover: - **AsySplat: Efficient Asymmetric 3D Gaussian Splatting for Long-Sequence Scene Modeling** (Hugging Face Daily)…
UnzipOn Locality and Length Generalization in Visual Reasoning
## Episode Summary In this episode, we cover: - **On Locality and Length Generalization in Visual Reasoning** (Hugging Face Daily) - [Read…
UnzipVIABench: A Comprehensive Video Benchmark Collected from Blind Individuals for Visual Impairment Assistance
## Episode Summary In this episode, we cover: - **VIABench: A Comprehensive Video Benchmark Collected from Blind Individuals for Visual Impairment Assistance**…
UnzipFrom Noisy Traces to Root Causes: Structural Trajectory Analysis and Causal Extraction for Agent Optimization
## Episode Summary In this episode, we cover: - **From Noisy Traces to Root Causes: Structural Trajectory Analysis and Causal Extraction for Agent…
UnzipTowards Autonomous and Auditable Medical Imaging Model Development
## Episode Summary In this episode, we cover: - **Towards Autonomous and Auditable Medical Imaging Model Development** (Hugging Face Daily) - [Read…
UnzipMetacognition in LLMs: Foundations, Progress, and Opportunities
## Episode Summary In this episode, we cover: - **Metacognition in LLMs: Foundations, Progress, and Opportunities** (Hugging Face Daily) - [Read…
UnzipMedPMC: A Systematic Framework for Scaling High-Fidelity Medical Multimodal Data for Foundation Models
## Episode Summary In this episode, we cover: - **MedPMC: A Systematic Framework for Scaling High-Fidelity Medical Multimodal Data for Foundation Models**…
UnzipLinear Attention Architectures: Mechanisms, Trade-offs, and Cross-Layer Routing
## Episode Summary In this episode, we cover: - **Linear Attention Architectures: Mechanisms, Trade-offs, and Cross-Layer Routing** (Hugging Face Daily) -…
UnzipDrugGen 2: A disease-aware language model for enhancing drug discovery
## Episode Summary In this episode, we cover: - **DrugGen 2: A disease-aware language model for enhancing drug discovery** (Hugging Face Daily) - [Read…
UnzipLongE2V: Long-Horizon Event-based Video Reconstruction, Prediction, and Frame Interpolation with Video Diffusion Models
## Episode Summary In this episode, we cover: - **LongE2V: Long-Horizon Event-based Video Reconstruction, Prediction, and Frame Interpolation with Video…
UnzipRoboTALES: Learning Reasoning-Guided Robot Policies via Task-Aligned Simulated Futures
## Episode Summary In this episode, we cover: - **RoboTALES: Learning Reasoning-Guided Robot Policies via Task-Aligned Simulated Futures** (Hugging Face Daily)…
UnzipFreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference
## Episode Summary In this episode, we cover: - **FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference**…
UnzipUnified Audio Intelligence Without Regressing on Text Intelligence
## Episode Summary In this episode, we cover: - **Unified Audio Intelligence Without Regressing on Text Intelligence** (Hugging Face Daily) - [Read…
UnzipVLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon
## Episode Summary In this episode, we cover: - **VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon** (Hugging Face Daily) -…
UnzipVisually Grounded Self-Reflection for Vision-Language Models via Reinforcement Learning
## Episode Summary In this episode, we cover: - **Visually Grounded Self-Reflection for Vision-Language Models via Reinforcement Learning** (arXiv) - [Read…
UnzipAutoMem: Automated Learning of Memory as a Cognitive Skill
## Episode Summary In this episode, we cover: - **AutoMem: Automated Learning of Memory as a Cognitive Skill** (Hugging Face Daily) - [Read…
UnzipEvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
## Episode Summary In this episode, we cover: - **EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments** (Hugging Face Daily) -…
UnzipTheoria: Rewrite-Acceptability Verification over Informal Reasoning States
## Episode Summary In this episode, we cover: - **Theoria: Rewrite-Acceptability Verification over Informal Reasoning States** (arXiv) - [Read…
UnzipHierarchical Experimentalist Agents
## Episode Summary In this episode, we cover: - **Hierarchical Experimentalist Agents** (Hugging Face Daily) - [Read…
UnzipBeyond IID: How General Are Tabular Foundation Models, Really?
## Episode Summary In this episode, we cover: - **Beyond IID: How General Are Tabular Foundation Models, Really?** (Hugging Face Daily) - [Read…
UnzipPerceptionRubrics: Calibrating Multimodal Evaluation to Human Perception
## Episode Summary In this episode, we cover: - **PerceptionRubrics: Calibrating Multimodal Evaluation to Human Perception** (arXiv) - [Read…
UnzipWhen Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
## Episode Summary In this episode, we cover: - **When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents…
UnzipPaying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models
## Episode Summary In this episode, we cover: - **Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models** (arXiv) - [Read…
UnzipGUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents
## Episode Summary In this episode, we cover: - **GUI vs. CLI: Execution Bottlenecks in Screen-Only and Skill-Mediated Computer-Use Agents** (Hugging Face…
UnzipConstraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints
## Episode Summary In this episode, we cover: - **Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output…
UnzipEscaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning
## Episode Summary In this episode, we cover: - **Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning**…