Computer-using agents (CUAs) are becoming increasingly capable; however, it remains difficult to scale evaluation of whether a trajectory truly fulfills a user instruction. In this work, we study reward modeling from execution video: a sequence of keyframes from an agent trajectory that is independent of the agent’s internal reasoning or actions. Although video-execution modeling is method-agnostic, it presents key challenges, including highly redundant layouts and subtle, localized cues that determine success. We introduce Execution Video Reward 53k (ExeVR-53k), a dataset of 53k high-quality video–task–reward triplets. We further propose adversarial instruction translation to synthesize negative samples with step-level annotations. To enable learning from long, high-resolution execution videos, we design spatiotemporal token pruning, which removes homogeneous regions and persistent tokens while preserving decisive UI changes. Building on these components, we fine-tune an Execution Video Reward Model (ExeVRM) that takes only a user instruction and a video-execution sequence to predict task success. Our ExeVRM 8B achieves 84.7% accuracy and 87.7% recall on video-execution assessment, outperforming strong proprietary models such as GPT-5.2 and Gemini-3 Pro across Ubuntu, macOS, Windows, and Android, while providing more precise temporal attribution. These results show that video-execution reward modeling can serve as a scalable, model-agnostic evaluator for CUAs.<\/p>\n","protected":false},"excerpt":{"rendered":"
Computer-using agents (CUAs) are becoming increasingly capable; however, it remains difficult to scale evaluation of whether a trajectory truly fulfills a user instruction. In this work, we study reward modeling from execution video: a sequence of keyframes from an agent trajectory that is independent of the agent’s internal reasoning or actions. Although video-execution modeling is […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":[{"type":"text","value":"Lin-Xin Song","user_id":0},{"type":"text","value":"Jieyu Zhang","user_id":0},{"type":"text","value":"Huan-Xin Sheng","user_id":0},{"type":"user_nicename","value":"Taiwei Shi","user_id":"44301"},{"type":"text","value":"G. Rahul","user_id":0},{"type":"user_nicename","value":"Yang Liu","user_id":"39594"},{"type":"text","value":"Ranjay Krishna","user_id":0},{"type":"text","value":"Jian Kang","user_id":0},{"type":"text","value":"Jie-Yu Zhao","user_id":0}],"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"arXiv.org","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"","msr_journal":"ArXiv","msr_number":"","msr_organization":"","msr_pages_string":"","msr_page_range_start":"","msr_page_range_end":"","msr_series":"","msr_volume":"abs\/2603.10178","msr_copyright":"","msr_conference_name":"","msr_doi":"10.48550\/arXiv.2603.10178","msr_arxiv_id":"2603.10178","msr_mag_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_release_tracker_id":"","msr_highlight_type":"","msr_date_display_format":"","msr_main_download_label":"","msr_external_link_label":"","msr_doi_label":"","msr_published_date":"2026-03-10","msr_startdate":"","msr_presentation_date":"","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_year":2026,"msr_month":3,"msr_day":10,"msr_microsoftintellectualproperty":false,"msr_pub_id":"1e898ee0ab5821caefb7d4d2c089d72272f27934","msr_publication_uploader":[{"type":"url","title":"https:\/\/arxiv.org\/abs\/2603.10178","label_id":243109,"id":false,"viewUrl":false},{"type":"url","title":"https:\/\/dblp.org\/rec\/journals\/corr\/abs-2603-10178.html","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_original_fields_of_study":[],"msr_s2_paper_id":"1e898ee0ab5821caefb7d4d2c089d72272f27934","msr_s2_pdf_url":"","msr_citation_count_updated":"","msr_citation_count":1,"msr_influential_citations":0,"msr_reference_count":73,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[{"provider":"s2","id":"1e898ee0ab5821caefb7d4d2c089d72272f27934"},{"provider":"doi","id":"10.48550\/arXiv.2603.10178"},{"provider":"arxiv","id":"2603.10178"},{"provider":"dblp","id":"journals\/corr\/abs-2603-10178"},{"provider":"corpusid","id":"286457391"}],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13562],"msr-publication-type":[193715],"msr-publisher":[],"msr-publication-cta":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[246691,263185],"msr-conference":[],"msr-journal":[268188],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1188308","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-computer-vision","msr-locale-en_us","msr-field-of-study-computer-science","msr-field-of-study-computer-vision-and-pattern-recognition"],"msr_publishername":"","msr_edition":"arXiv.org","msr_affiliation":"","msr_published_date":"2026-03-10","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"ArXiv","msr_volume":"abs\/2603.10178","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":0,"msr_main_download":"","msr_publicationurl":"","msr_doi":"10.48550\/arXiv.2603.10178","msr_publication_uploader":[{"type":"url","title":"https:\/\/arxiv.org\/abs\/2603.10178","label_id":243109,"id":false,"viewUrl":false},{"type":"url","title":"https:\/\/dblp.org\/rec\/journals\/corr\/abs-2603-10178.html","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_citation_count":1,"msr_citation_count_updated":"","msr_s2_paper_id":"1e898ee0ab5821caefb7d4d2c089d72272f27934","msr_influential_citations":0,"msr_reference_count":73,"msr_arxiv_id":"2603.10178","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Lin-Xin Song","user_id":0,"rest_url":false},{"type":"text","value":"Jieyu Zhang","user_id":0,"rest_url":false},{"type":"text","value":"Huan-Xin Sheng","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Taiwei Shi","user_id":44301,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Taiwei Shi"},{"type":"text","value":"G. Rahul","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Yang Liu","user_id":39594,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Yang Liu"},{"type":"text","value":"Ranjay Krishna","user_id":0,"rest_url":false},{"type":"text","value":"Jian Kang","user_id":0,"rest_url":false},{"type":"text","value":"Jie-Yu Zhao","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[199560,1012650],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"article","related_content":[],"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188308","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188308\/revisions"}],"predecessor-version":[{"id":1189003,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1188308\/revisions\/1189003"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1188308"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1188308"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1188308"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1188308"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1188308"},{"taxonomy":"msr-publication-cta","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-cta?post=1188308"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1188308"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1188308"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1188308"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1188308"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1188308"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1188308"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1188308"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1188308"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}