Skip to content

AI论文速递 2026年04月28日（HuggingFace Daily Papers）

数据来源：https://huggingface.co/papers 采集时间：2026-04-28

📌 重点关注

  1. OpenMobile: Building Open Mobile Agents with Task and Trajectory Synthesis | arXiv — 【重点关注】 Mobile agents powered by vision-language models have demonstrated impressive ...
  2. EmbodiedMidtrain: Bridging the Gap between Vision-Language Models and Vision-Language-Action Models via Mid-training | arXiv — 【重点关注】 Vision-Language-Action Models (VLAs) inherit their visual and linguistic capa...
  3. WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning | arXiv — 【重点关注】 While Large Language Models (LLMs) excel at function-level code generation, p...

📋 其他值得关注

  1. LLaTiSA: Towards Difficulty-Stratified Time Series Reasoning from Visual Perception to Semantics | arXiv — Comprehensive understanding of time series remains a significant challenge fo...
  2. PersonalAI: A Systematic Comparison of Knowledge Graph Storage and Retrieval Approaches for Personalized LLM agents | arXiv — Personalizing language models by effectively incorporating user interaction h...
  3. Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Tasks | arXiv — Long horizon interactive environments are a testbed for evaluating agents ski...
  4. 3D-VCD: Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive Decoding | arXiv — Large multimodal models are increasingly used as the reasoning core of embodi...
  5. Rewarding the Scientific Process: Process-Level Reward Modeling for Agentic Data Analysis | arXiv — Process Reward Models (PRMs) have achieved remarkable success in augmenting t...
  6. Benign Fine-Tuning Breaks Safety Alignment in Audio LLMs | arXiv — Prior work shows that fine-tuning aligned models on benign data degrades safe...
  7. AgriIR: A Scalable Framework for Domain-Specific Knowledge Retrieval | arXiv — This paper introduces AgriIR, a configurable retrieval augmented generation (...