OpenDataLab · 9 records

dateidtitleauthorsvenue↑
2026-08-142608.10538SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language ModelsChenhao Dang et al.arXiv9
2026-07-072607.04033OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern OptimizersSiyuan Li et al.arXiv49
2026-06-242606.24133Holistic Data Scheduler for LLM Pre-training via Multi-Objective Reinforcement LearningChenhao Dang et al.arXiv12
2026-06-242606.24112ReMMD: Realistic Multilingual Multi-Image Agentic Verification for Multimodal Misinformation DetectionChenhao Dang et al.arXiv5
2026-06-232505.23878AC-ODM: Actor--Critic Online Data Mixing for Sample-Efficient LLM PretrainingJing Ma et al.arXiv1
2026-06-082606.07454PaperFlow: Profiling, Recommending, and Adapting Across Daily Paper StreamsFuqiang Wang et al.arXiv14
2026-05-182605.12882CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document IntelligenceDongsheng Ma et al.arXiv27
2026-05-182605.15963PAGER: Bridging the Semantic-Execution Gap in Point-Precise Geometric GUI ControlJingxuan Wei et al.arXiv13
2026-05-122605.10341PaperFit: Vision-in-the-Loop Typesetting Optimization for Scientific DocumentsBihui Yu et al.arXiv32