Editing Live Video, Shrinking VLMs to 2.7 Bits

Featured

• Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence score 7
  入选 HF Daily Papers; HF 热度: 4 upvotes (+1); 有代码实现; 关键词(1): open-source
• InfinityEdit: Infinite Video Editing with a Lightweight Edit-Ignition Adapter score 6
  入选 HF Daily Papers; 有代码实现; 关键词(1): lightweight
• AgentMercury: Your Agent Can Synthesize Verifiable Environments for Business Scenarios at scale score 5
  入选 HF Daily Papers; HF 热度: 2 upvotes (+1); 关键词(4): fine-tuning, tool use, coding, reasoning

Also Worth Noting

• Towards Faithful Simulation of Human Shopping Behavior score 4
  入选 HF Daily Papers; HF 热度: 2 upvotes (+1)
• Llama-Mobile: Efficient 2.7-Bit Quantization of VLMs score 4
  入选 HF Daily Papers; 关键词(2): quantization, vision-language
• CLEAR: Continuous Latent Adapter Routing for Utility-Preserving LLM Safety Alignment score 4
  入选 HF Daily Papers; 关键词(1): lightweight
• SPARC: Single-Pass Scaling for Motion Forecasting with Conformal Bayesian Last Layers score 4
  关键词(3): scaling, lightweight, deployment; 顶会接收: ECCV
• Identify, Locate, Link: End-to-End Key-Value Extraction from Document Images score 4
  机构: ETH Zurich; 关键词(2): fine-tune, vision-language
• Target-Aware Calibration Data Selection for Preserving Uncertainty in Quantized Language Models score 4
  关键词(4): lightweight, compression, quantization, deployment; 顶会接收: EMNLP
• Scaling Unsupervised Word Alignment to Documents via Structural Constraints score 4
  关键词(2): scaling, lightweight; 顶会接收: EMNLP
• ReFrame: Evidence-Guided Test-Time Safety Alignment in Multimodal Large Language Models score 4
  关键词(3): lightweight, reasoning, jailbreak; 顶会接收: EMNLP
• Is Visual Prompting All You Need? Studying VLM Spatial Reasoning under Progressive Visual Scaffolds score 4
  关键词(4): lightweight, GRPO, reasoning, vision-language; 顶会接收: EMNLP
• Benchmarking Patent Drafting from Inventor-Style Disclosures score 4
  关键词(1): open-source; 顶会接收: EMNLP
• Re$^3$Cap: Retrieval-Guided Refinement for Image Captioning Enhancement via Reinforcement Learning score 4
  关键词(4): fine-tuning, GRPO, reasoning, vision-language; 顶会接收: EMNLP
• MV2GF: Multi-view Pedestrian Detection with a Visual Geometric Foundation Model score 3
  顶会接收: ECCV
• Natural-Language-Guided Generator-Agnostic Shortlisting for Protein Binder Design score 3
  顶会接收: ICML
• Ontology-Driven Structural Regularization for Document-Level Relation Extraction score 3
  顶会接收: EMNLP
• ForeDreamer: A Self-Evolving Dual-Agent Memory Architecture for Future Event Prediction score 3
  顶会接收: EMNLP