BEGIN:VCALENDAR VERSION:2.0 PRODID:-//swoogo.com//NONSGML kigkonsult.se iCalcreator 2.41.90// CALSCALE:GREGORIAN UID:34333435-3433-4362-b164-343536623633 BEGIN:VEVENT UID:faccfb28e5ed6a5bc9031f8680ed049eed1b8633@swoogo.com DTSTAMP:20261012T093929Z DESCRIPTION:Retrieval-augmented generation (RAG) is widely used to enhance AI’s knowledge with new or changing information\, but real-world complexit ies challenge its reliability. This talk explores how Large Language Model s (LLMs) process retrieved evidence\, why they often fail to utilize it co rrectly\, and how misinformation risks emerge from unreliable or insuffici ent context. Based on recent research\, I will highlight key pitfalls in L LM-driven fact verification and discuss strategies for improving fact chec king robustness. DTSTART:20251105T134500Z DTEND:20251105T141500Z LAST-MODIFIED:20261012T093929Z LOCATION: SEQUENCE:0 STATUS:CONFIRMED SUMMARY:Reality Check of LLM-driven Fact Verification: Retrieving and Utili sing Evidence in the Wild TRANSP:OPAQUE X-ALT-DESC;FMTTYPE=text/html:

Retrieval-augmented generation (RAG) is wid ely used to enhance AI’s knowledge with new or changing information\, but real-world complexities challenge its reliability. This talk explores how Large Language Models (LLMs) process retrieved evidence\, why they often f ail to utilize it correctly\, and how misinformation risks emerge from unr eliable or insufficient context. Based on recent research\, I will highlig ht key pitfalls in LLM-driven fact verification and discuss strategies for improving fact checking robustness.

BEGIN:VALARM UID:30663038-3530-4636-a333-326431343132 ACTION:DISPLAY DESCRIPTION:Retrieval-augmented generation (RAG) is widely used to enhance AI’s knowledge with new or changing information\, but real-world complexit ies challenge its reliability. This talk explores how Large Language Model s (LLMs) process retrieved evidence\, why they often fail to utilize it co rrectly\, and how misinformation risks emerge from unreliable or insuffici ent context. Based on recent research\, I will highlight key pitfalls in L LM-driven fact verification and discuss strategies for improving fact chec king robustness. TRIGGER:-PT15M END:VALARM END:VEVENT END:VCALENDAR