{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/tracevla-visual-trace-prompting-enhances","title":"TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies","arxiv_id":"2412.10345","date":"2024-12-13","proceeding":null,"authors":["Ruijie Zheng","Yongyuan Liang","Shuaiyi Huang","Jianfeng Gao","Hal Daumé III","Andrey Kolobov","Furong Huang","Jianwei Yang"],"abstract":"Although large vision-language-action (VLA) models pretrained on extensive robot datasets offer promising generalist policies for robotic learning, they still struggle with spatial-temporal dynamics in interactive robotics, making them less effective in handling complex tasks, such as manipulation. In this work, we introduce visual trace prompting, a simple yet effective approach to facilitate VLA models' spatial-temporal awareness for action prediction by encoding state-action trajectories visually. We develop a new TraceVLA model by finetuning OpenVLA on our own collected dataset of 150K robot manipulation trajectories using visual trace prompting. Evaluations of TraceVLA across 137 configurations in SimplerEnv and 4 tasks on a physical WidowX robot demonstrate state-of-the-art performance, outperforming OpenVLA by 10% on SimplerEnv and 3.5x on real-robot tasks and exhibiting robust generalization across diverse embodiments and scenarios. To further validate the effectiveness and generality of our method, we present a compact VLA model based on 4B Phi-3-Vision, pretrained on the Open-X-Embodiment and finetuned on our dataset, rivals the 7B OpenVLA baseline while significantly improving inference efficiency.","url_abs":"https://arxiv.org/abs/2412.10345v3","url_pdf":"https://arxiv.org/pdf/2412.10345v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"robot-manipulation","task_name":"Robot Manipulation"},{"task_slug":"vision-language-action","task_name":"Vision-Language-Action"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/robot-manipulation-on-simpler-env","task":"Robot Manipulation","dataset":"SimplerEnv-Google Robot","model":"TraceVLA","rank_in_archive_order":7,"of":9,"metrics":{"Variant Aggregation":"0.450","Variant Aggregation-Move Near":"0.564","Variant Aggregation-Open/Close Drawer":"0.310","Variant Aggregation-Pick Coke Can":"0.600","Visual Matching":"0.460","Visual Matching-Move Near":"0.600","Visual Matching-Open/Close Drawer":"0.240","Visual Matching-Pick Coke Can":"0.560"},"uses_additional_data":true}],"syntology":{"syntology_url":"https://syntology.ai/paper/2412.10345","atlas_url":"https://app.syntology.ai/?focus=2412.10345","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}