Hugging Face · Daily Papers
Daily machine learning papers highlighted by Hugging Face.
Read through RSSHubAppears in:IA
Latest from Hugging Face · Daily Papers
- Muir: Linear Attention with Sensitivity-Allocated Gating and Matched-Filter Readouthace 3 d
- Terminal-style Mathematical Modeling Bench: The First Agentic and Verifiable Benchmark for Mathematical Modelinghace 3 d
- SAE Interprets MoE: Sparse AutoEncoders Reveal That MoE Experts Exhibit Human-Interpretable Fine-Grained Specializationhace 3 d
- Towards Real-Time Full-Waveform LiDAR Transformers via Intensity-Guided Token Reduction and Physics-Aware Augmentationhace 3 d
- Reducing Representational Interference in Retrieval Embeddings with Trainable Prefix Tokenshace 4 d
- VFlash: Fast, Accurate Visual Planning via Flash Primitive Decodinghace 4 d
- SLA: Scaling Spatial-Language-Action Model via Continual Pre-traininghace 4 d
- WeLM-Image: A Unified Image Generation Model towards Visual Intelligencehace 4 d
- Towards Foundation Decision Models: Drexhace 4 d
- EDGAR: Efficient Diffusion-Driven Gaussian Avatar Refinement via Synthetic Failurehace 4 d
- SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimizationhace 4 d
- Information Games: Strategic Crowding and Firm Repositioning in Language-Model Spacehace 4 d
- NPCBench: A Clinical Apprenticeship Benchmark for Multimodal Care-Pathway Reasoning in Nasopharyngeal Carcinomahace 4 d
- On the Opportunities and Challenges of Prefill-to-Autoregressive Cross-Model Transferhace 4 d
- Signed but Steerable: Measuring and Closing the Text-Vector Integrity Gap in Offline Retrieval-Augmented Generationhace 4 d
- TailSteer: Trajectory-Tail Steering for Drift-Resistant Multi-turn Image Editinghace 5 d
- Open-HumanScale: An Open Recipe for Scaling Human and Robot Experience for Embodied Foundation Modelshace 5 d
- Future Representation Distillation for Imagination-free World Action Modelshace 5 d
- Decode Before You Hear: Speculative Listening for Chunk-Interleaved Full-Duplex Speech Language Modelshace 5 d
- Tools Need Not Speak in Text: Latent Context Compression for Agentshace 5 d
- From Autonomous Agents to Cognitive Symbionts: Toward AI-Augmented Human Intelligencehace 5 d
- Harness-S: Learning from Sibling Harnesses for Generalizable Agent Traininghace 6 d
- VisSearch-Bench: Evaluating Long Horizon Visual Search in Multimodal Agentshace 6 d
- StepX-OPD: Rethinking Multi-Teacher On-Policy Distillation in Diffusion Modelshace 6 d
- HeLM: Hierarchical Explicit Language Memory for Non-Markovian Manipulationhace 6 d
- EvoMemState: Evolving Memory State for Personal Agentshace 7 d
- Decoupled Predictive Decision Making: A Shared Conditional Predictor for Agentshace 8 d
- Random Is Almost Enough? Stable and Low-Cost Benchmark Compression via BestRandhace 9 d
- Hard Sample-Conditioned Decorrelated Attention for Robust Face Recognitionhace 9 d
- SynthFaces: Balanced Large Scale Human Datasethace 11 d