{"id":1146661,"date":"2025-08-05T09:00:07","date_gmt":"2025-08-05T16:00:07","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-video&#038;p=1146661"},"modified":"2025-08-06T13:42:45","modified_gmt":"2025-08-06T20:42:45","slug":"veritrail-detect-hallucination-and-trace-provenance-in-ai-workflows","status":"publish","type":"msr-video","link":"https:\/\/www.microsoft.com\/en-us\/research\/video\/veritrail-detect-hallucination-and-trace-provenance-in-ai-workflows\/","title":{"rendered":"VeriTrail: Detect hallucination and trace provenance in AI workflows"},"content":{"rendered":"\n<p>Dasha Metropolitansky, Research Data Scientist, Microsoft Research Special Projects, introduces VeriTrail, a new method for closed-domain hallucination detection in multi-step AI workflows. Unlike prior methods, VeriTrail provides traceability: it identifies where hallucinated content was likely introduced, and it establishes the provenance of faithful content by tracing a path to the source text. VeriTrail also outperforms baseline methods in hallucination detection. The combination of traceability and effective hallucination detection makes VeriTrail a powerful tool for auditing the integrity of content generated by language models.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/veritrail-closed-domain-hallucination-detection-with-traceability\/\">Microsoft Research Special Projects VeriTrail paper<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/blog\/veritrail-detecting-hallucination-and-tracing-provenance-in-multi-step-ai-workflows\/\">VeriTrail blog post<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/video\/claimify-extracting-high-quality-claims-from-language-model-outputs\/\">Claimify video<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/towards-effective-extraction-and-evaluation-of-factual-claims\/\">Claimify paper<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/blog\/claimify-extracting-high-quality-claims-from-language-model-outputs\/\">Claimify blog post<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Dasha Metropolitansky, Research Data Scientist, Microsoft Research Special Projects, introduces VeriTrail, a new method for closed-domain hallucination detection in multi-step AI workflows. Unlike prior methods, VeriTrail provides traceability: it identifies where hallucinated content was likely introduced, and it establishes the provenance of faithful content by tracing a path to the source text. VeriTrail also outperforms [&hellip;]<\/p>\n","protected":false},"featured_media":1146664,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_hide_image_in_river":null,"footnotes":""},"research-area":[13556,13545],"msr-video-type":[],"msr-locale":[268875],"msr-post-option":[269148,269142],"msr-session-type":[],"msr-impact-theme":[],"msr-pillar":[],"msr-episode":[],"msr-research-theme":[],"class_list":["post-1146661","msr-video","type-msr-video","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-research-area-human-language-technologies","msr-locale-en_us","msr-post-option-approved-for-river","msr-post-option-include-in-river"],"msr_download_urls":"","msr_external_url":"https:\/\/youtu.be\/pi6Km-vKSVY","msr_secondary_video_url":"","msr_video_file":"http:\/\/0","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/1146661","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-video"}],"version-history":[{"count":5,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/1146661\/revisions"}],"predecessor-version":[{"id":1146676,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/1146661\/revisions\/1146676"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/1146664"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1146661"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1146661"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=1146661"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1146661"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1146661"},{"taxonomy":"msr-session-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-session-type?post=1146661"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1146661"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1146661"},{"taxonomy":"msr-episode","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-episode?post=1146661"},{"taxonomy":"msr-research-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-theme?post=1146661"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}