{"id":1179033,"date":"2026-01-01T00:00:00","date_gmt":"2026-01-01T08:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-research-item&#038;p=1179033"},"modified":"2026-07-19T10:58:53","modified_gmt":"2026-07-19T17:58:53","slug":"can-ai-revise-research-papers-with-human-review-feedback-an-empirical-study-and-benchmark","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/can-ai-revise-research-papers-with-human-review-feedback-an-empirical-study-and-benchmark\/","title":{"rendered":"Can AI Revise Research Papers with Human Review Feedback? An Empirical Study and Benchmark"},"content":{"rendered":"\n\n\n<p class=\"wp-block-paragraph\">The rise of Human-AI collaboration can effectively speed up the research process for experts and allow anyone with critical thinking skills to conduct innovative work. A key part of this collaboration is the AI\u2019s ability to improve a paper<br>with human feedback\u2014updating both the text and experiments to meet high standards. To evaluate this skill, we introduce ReviseBench, an extensible benchmark built on real academic data that can be easily scaled via agent-driven automated data collection. It tests the skills of Large Language Models (LLMs) on paper in terpretation, experimental implementation, and paper formulation, using authors\u2019 camera-ready versions as natural human baselines. To facilitate a fine-grained assessment, we further propose ReviseArena, a platform supporting pair-wise comparisons between different AI revised papers. Our initial evaluation results on ReviseBench reveal that even state-of-the art foundation LLMs struggle significantly in this domain, achieving a win rate of less than 10% against human experts, and facing issues like incremental revision, unprofessional revision, and potential data fabrication. Our code and data are released publicly at: <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/github.com\/CGCL-codes\/ReviseBench.\">https:\/\/github.com\/CGCL-codes\/ReviseBench.<span class=\"sr-only\"> (opens in new tab)<\/span><\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n","protected":false},"excerpt":{"rendered":"<p>The rise of Human-AI collaboration can effectively speed up the research process for experts and allow anyone with critical thinking skills to conduct innovative work. A key part of this collaboration is the AI\u2019s ability to improve a paperwith human feedback\u2014updating both the text and experiments to meet high standards. To evaluate this skill, we [&hellip;]<\/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":"Zihan Luo","user_id":0},{"type":"text","value":"Hong Huang","user_id":0},{"type":"user_nicename","value":"Jianxun Lian","user_id":"38470"},{"type":"text","value":"Yu Chang","user_id":0},{"type":"user_nicename","value":"Xing Xie","user_id":"34906"},{"type":"text","value":"Hai Jin","user_id":0}],"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"","msr_journal":"","msr_number":"","msr_organization":"","msr_pages_string":"","msr_page_range_start":"","msr_page_range_end":"","msr_series":"","msr_volume":"","msr_copyright":"","msr_conference_name":"","msr_doi":"10.18653\/v1\/2026.findings-acl.887","msr_arxiv_id":"","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":"2026-01-01","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":1,"msr_day":1,"msr_microsoftintellectualproperty":false,"msr_pub_id":"81cc094c5a8b29f41fdb10642aabf2f490a0fe72","msr_publication_uploader":[{"type":"url","title":"https:\/\/aclanthology.org\/2026.findings-acl.887.pdf","label_id":243132,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_original_fields_of_study":[],"msr_s2_paper_id":"","msr_s2_pdf_url":"","msr_citation_count_updated":"","msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":0,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[{"provider":"s2","id":"81cc094c5a8b29f41fdb10642aabf2f490a0fe72"},{"provider":"doi","id":"10.18653\/v1\/2026.findings-acl.887"},{"provider":"corpusid","id":"289820377"}],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13554],"msr-publication-type":[270373],"msr-publisher":[],"msr-publication-cta":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[270445],"msr-journal":[270444],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1179033","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-human-computer-interaction","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2026-01-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","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":"10.18653\/v1\/2026.findings-acl.887","msr_publication_uploader":[{"type":"url","title":"https:\/\/aclanthology.org\/2026.findings-acl.887.pdf","label_id":243132,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"","msr_influential_citations":0,"msr_reference_count":0,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Zihan Luo","user_id":0,"rest_url":false},{"type":"text","value":"Hong Huang","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Jianxun Lian","user_id":38470,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jianxun Lian"},{"type":"text","value":"Yu Chang","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Xing Xie","user_id":34906,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Xing Xie"},{"type":"text","value":"Hai Jin","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[199560],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"misc","related_content":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1179033","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":3,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1179033\/revisions"}],"predecessor-version":[{"id":1179036,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1179033\/revisions\/1179036"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1179033"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1179033"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1179033"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1179033"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1179033"},{"taxonomy":"msr-publication-cta","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-cta?post=1179033"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1179033"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1179033"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1179033"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1179033"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1179033"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1179033"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1179033"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1179033"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}