{"arxiv_id":"2608.22301","incoming":{"papers":0,"sentences":0,"groups":[],"unlabelled_sentences":0},"outgoing":{"papers":19,"sentences":31,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":8,"sentences":8,"quotes":[{"arxiv_id":"2504.16054","title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","url":"/paper/2504.16054","section":"Baseline adapters","sentence":"VLA models (GR00T-N1.6 [ref] , RDT-1B [ref] , $π_0.5$ [ref] , OpenVLA [ref] ) are language-conditioned, so the fixed captioner $T(V)$ of Section [ref] supplies the prompt.","intent_p":0.96,"contests_p":null},{"arxiv_id":"2406.09246","title":"OpenVLA: An Open-Source Vision-Language-Action Model","url":"/paper/2406.09246","section":"Baseline adapters","sentence":"VLA models (GR00T-N1.6 [ref] , RDT-1B [ref] , $π_0.5$ [ref] , OpenVLA [ref] ) are language-conditioned, so the fixed captioner $T(V)$ of Section [ref] supplies the prompt.","intent_p":0.96,"contests_p":null},{"arxiv_id":"2503.14734","title":"GR00T N1: An Open Foundation Model for Generalist Humanoid Robots","url":"/paper/2503.14734","section":"Baseline adapters","sentence":"VLA models (GR00T-N1.6 [ref] , RDT-1B [ref] , $π_0.5$ [ref] , OpenVLA [ref] ) are language-conditioned, so the fixed captioner $T(V)$ of Section [ref] supplies the prompt.","intent_p":0.96,"contests_p":null},{"arxiv_id":"2502.14786","title":"SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features","url":"/paper/2502.14786","section":"Baseline adapters","sentence":"Vision-action models (ACT [ref] , Diffusion Policy [ref] , VQ-BeT [ref] ) gain video conditioning by concatenating a task embedding from one of three frozen encoders (DINOv2-ViT-L/14 [ref] , SigLIP2-SO400M [ref] , VideoMAE-Large [ref] ) to the observation tokens.","intent_p":0.88,"contests_p":null},{"arxiv_id":"2304.07193","title":"DINOv2: Learning Robust Visual Features without Supervision","url":"/paper/2304.07193","section":"Baseline adapters","sentence":"Vision-action models (ACT [ref] , Diffusion Policy [ref] , VQ-BeT [ref] ) gain video conditioning by concatenating a task embedding from one of three frozen encoders (DINOv2-ViT-L/14 [ref] , SigLIP2-SO400M [ref] , VideoMAE-Large [ref] ) to the observation tokens.","intent_p":0.88,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":17,"sentences":17,"quotes":[{"arxiv_id":"2510.07773","title":null,"url":"/paper/2510.07773","section":"Related Works","sentence":"Prior work learns from human videos by retargeting extracted hand or object poses [ref] , following optical-flow or object-centric motion cues [ref] , inpainting human hands as robot grippers [ref] , or filtering cross-embodiment actions [ref] .","intent_p":0.96,"contests_p":null},{"arxiv_id":"2505.07096","title":null,"url":"/paper/2505.07096","section":"Related Works","sentence":"Prior work learns from human videos by retargeting extracted hand or object poses [ref] , following optical-flow or object-centric motion cues [ref] , inpainting human hands as robot grippers [ref] , or filtering cross-embodiment actions [ref] .","intent_p":0.96,"contests_p":null},{"arxiv_id":"2508.09976","title":"Masquerade: Learning from In-the-wild Human Videos using Data-Editing","url":"/paper/2508.09976","section":"Related Works","sentence":"Prior work learns from human videos by retargeting extracted hand or object poses [ref] , following optical-flow or object-centric motion cues [ref] , inpainting human hands as robot grippers [ref] , or filtering cross-embodiment actions [ref] .","intent_p":0.96,"contests_p":null},{"arxiv_id":"2202.10448","title":"Robotic Telekinesis: Learning a Robotic Hand Imitator by Watching Humans on Youtube","url":"/paper/2202.10448","section":"Related Works","sentence":"Prior work learns from human videos by retargeting extracted hand or object poses [ref] , following optical-flow or object-centric motion cues [ref] , inpainting human hands as robot grippers [ref] , or filtering cross-embodiment actions [ref] .","intent_p":0.96,"contests_p":null},{"arxiv_id":"2505.08787","title":null,"url":"/paper/2505.08787","section":"Related Works","sentence":"Skill-representation methods such as XSkill [ref] , UniSkill [ref] , RHyME [ref] , and ImMimic [ref] learn human-robot correspondences, and Vid2Robot [ref] conditions control on a video prompt.","intent_p":0.83,"contests_p":null}]}],"unlabelled_sentences":6},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2608.22301"}}