BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/New_York BEGIN:DAYLIGHT DTSTART:20260308T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20261101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260615T194726Z UID:E8EAE8AF-E75F-4CDF-9343-4AF53AAE7E90 DTSTART;TZID=America/New_York:20260714T180000 DTEND;TZID=America/New_York:20260714T200000 DESCRIPTION:The Self-Correcting RAG:\nImproving Retrieval via User Feedback Loops\n\n[REGISTER](https://forms.cloud.microsoft/Pages/ResponsePage.aspx ?id=v4j5cvGGr0GRqy180BHbR-iqRQAG1WhNvYrCKGtVBKBUNDlWSUtXMDVYTEpYNDQ1QTFXOE pWV1A1Ri4u)\n\nYour RAG pipeline was built to surface the right answers — so why does it keep retrieving the wrong documents?\n\nStandard Retrie val-Augmented Generation works well out of the box\, but in specialized te chnical domains\, general-purpose embeddings struggle with internal termin ology\, jargon\, and nuanced context. The conventional fix — fine-tuning models or rebuilding embeddings — is expensive\, slow\, and breaks down the moment your knowledge base changes.\n\nThere's a better way. In this workshop\, we'll introduce FLAIR (Feedback Learning for Adaptive Informati on Retrieval)\, a lightweight framework that continuously adapts your retr ieval strategy using domain-expert feedback — no model re-training requi red. You'll learn how FLAIR:\n\n- Gathers real and LLM-synthesized query i ndicators offline to understand what experts actually need\n- Deploys a tw o-track online ranking system that dynamically promotes relevant documents and filters out past retrieval errors — in real time\n- Evolves with yo ur knowledge base so retrieval quality improves continuously\, not just af ter the next expensive rebuild\n\n[REGISTER](https://forms.cloud.microsoft /Pages/ResponsePage.aspx?id=v4j5cvGGr0GRqy180BHbR-iqRQAG1WhNvYrCKGtVBKBUND lWSUtXMDVYTEpYNDQ1QTFXOEpWV1A1Ri4u)\nDue to Security\, no walk-ins will be accepted. All attendees must register and log-in.\n[]\nSPONSORED BY:\nIEE E Computational Intelligence Society Boston Chapter and the IEEE Boston AI Local Group\n\nCo-sponsored by: IEEE-Boston AI Local Group\n\nSpeaker(s): Nutan Sahoo\, \n\nRoom: NERD/1250\, Abigail Adams\, Microsoft NERD Center \, Kendall Square\, 1 Memorial Drive\, Cambridge\, Massachusetts\, United States\, 02142 LOCATION:Room: NERD/1250\, Abigail Adams\, Microsoft NERD Center\, Kendall Square\, 1 Memorial Drive\, Cambridge\, Massachusetts\, United States\, 02 142 ORGANIZER:info@voptronix.com SEQUENCE:232 SUMMARY:The Self-Correcting RAG: Improving Retrieval via User Feedback Loop s URL;VALUE=URI:https://events.vtools.ieee.org/m/560479 X-ALT-DESC:Description: <br /><div style="text-align: center\;">\n<div styl e="vertical-align: top\; position: relative\; display: inline-block\; widt h: 100%\; min-width: 320px\; max-width: 600px\; background-color: rgb(255\ , 255\, 255)\;">\n<div style="display: block\; margin: 10px 20px 10px 20px \;">\n<div style="text-align: center\;"><span style="font-size: 20pt\; fon t-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; font-we ight: bold\;">The Self-Correcting RAG:</span><span style="font-size: 18pt\ ; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; fo nt-weight: bold\;"><br></span><span style="font-size: 16pt\; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; font-weight: bold \;">Improving Retrieval via User Feedback Loops<br><br></span>\n<table sty le="box-sizing: border-box\; vertical-align: bottom\; position: relative\; display: inline-table\; width: 50%\; height: 30px\; min-width: 100px\; ma x-width: 100px\; background-color: rgb(255\, 255\, 255)\; border-image-sou rce: none\; border-image-slice: 100%\; border-image-width: 1\; border-imag e-outset: 0\; border-image-repeat: stretch\; table-layout: fixed\; border- width: 1px\; border-color: rgb(32\, 178\, 227)\; border-style: solid\;" ce llspacing="0" cellpadding="3">\n<tbody>\n<tr>\n<td style="vertical-align: top\; overflow: hidden\; border-image-source: none\; border-image-slice: 1 00%\; border-image-width: 1\; border-image-outset: 0\; border-image-repeat : stretch\; border-width: 1px\; border-color: rgb(32\, 178\, 227)\; border -style: solid\;" width="100%" height="20px">\n<div id="cell_75690335">\n<d iv style="text-align: center\;"><span style="font-size: 12pt\; font-family : Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; font-weight: bo ld\; line-height: 1.3\;"><a style="color: rgb(32\, 178\, 227)\;" href="htt ps://forms.cloud.microsoft/Pages/ResponsePage.aspx?id=v4j5cvGGr0GRqy180BHb R-iqRQAG1WhNvYrCKGtVBKBUNDlWSUtXMDVYTEpYNDQ1QTFXOEpWV1A1Ri4u" target="_bla nk" rel="noopener">REGISTER</a></span></div>\n</div>\n</td>\n</tr>\n</tbod y>\n</table>\n<span style="font-size: 10pt\; font-family: Arial\, Helvetic a\, sans-serif\; color: rgb(0\, 0\, 0)\;"><br></span><span style="font-siz e: 16pt\; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; font-weight: bold\;"><br></span></div>\n<div style="text-align: left \;"><span style="font-size: 16pt\; font-family: Arial\, Helvetica\, sans-s erif\; color: rgb(0\, 0\, 0)\; font-weight: bold\;">Your RAG pipeline was built to surface the right answers &mdash\; so why does it keep retrieving the wrong documents?</span><span style="font-size: 12pt\; font-family: Ar ial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\;"><br></span><span s tyle="font-size: 12pt\; font-family: Arial\, Helvetica\, sans-serif\; colo r: rgb(0\, 0\, 0)\; line-height: 1.3\;"><br>Standard Retrieval-Augmented G eneration works well out of the box\, but in specialized technical domains \, general-purpose embeddings struggle with internal terminology\, jargon\ , and nuanced context. The conventional fix &mdash\; fine-tuning models or rebuilding embeddings &mdash\; is expensive\, slow\, and breaks down the moment your knowledge base changes.<br></span><span style="font-size: 12pt \; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; f ont-weight: bold\; line-height: 1.3\;"><br>There's a better way. In this w orkshop\, we'll introduce FLAIR</span><span style="font-size: 12pt\; font- family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; line-heig ht: 1.3\;"> (Feedback Learning for Adaptive Information Retrieval)\, a lig htweight framework that continuously adapts your retrieval strategy using domain-expert feedback &mdash\; no model re-training required. You'll lear n how FLAIR:<br><br></span>\n<ul style="font-size: 12pt\; color: rgb(0\, 0 \, 0)\; margin: 0px\;">\n<li><span style="font-size: 12pt\; font-family: A rial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; line-height: 1.3\; ">Gathers real and LLM-synthesized query indicators offline to understand what experts actually need</span></li>\n<li><span style="font-size: 12pt\; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; lin e-height: 1.3\;">Deploys a two-track online ranking system that dynamicall y promotes relevant documents and filters out past retrieval errors &mdash \; in real time</span></li>\n<li><span style="font-size: 12pt\; font-famil y: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; line-height: 1 .3\;">Evolves with your knowledge base so retrieval quality improves conti nuously\, not just after the next expensive rebuild</span></li>\n</ul>\n<s pan style="font-size: 12pt\; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\; line-height: 1.3\;"><br></span></div>\n<div style ="text-align: center\;">\n<table style="box-sizing: border-box\; vertical- align: bottom\; position: relative\; display: inline-table\; width: 50%\; height: 30px\; min-width: 100px\; max-width: 100px\; background-color: rgb (255\, 255\, 255)\; border-image-source: none\; border-image-slice: 100%\; border-image-width: 1\; border-image-outset: 0\; border-image-repeat: str etch\; table-layout: fixed\; border-width: 1px\; border-color: rgb(32\, 17 8\, 227)\; border-style: solid\;" cellspacing="0" cellpadding="3">\n<tbody >\n<tr>\n<td style="vertical-align: top\; overflow: hidden\; border-image- source: none\; border-image-slice: 100%\; border-image-width: 1\; border-i mage-outset: 0\; border-image-repeat: stretch\; border-width: 1px\; border -color: rgb(32\, 178\, 227)\; border-style: solid\;" width="100%" height=" 20px">\n<div id="cell_24ee9737">\n<div style="text-align: center\;"><span style="font-size: 12pt\; font-family: Arial\, Helvetica\, sans-serif\; col or: rgb(0\, 0\, 0)\; font-weight: bold\; line-height: 1.3\;"><a style="col or: rgb(32\, 178\, 227)\;" href="https://forms.cloud.microsoft/Pages/Respo nsePage.aspx?id=v4j5cvGGr0GRqy180BHbR-iqRQAG1WhNvYrCKGtVBKBUNDlWSUtXMDVYTE pYNDQ1QTFXOEpWV1A1Ri4u" target="_blank" rel="noopener">REGISTER</a></span> </div>\n</div>\n</td>\n</tr>\n</tbody>\n</table>\n<span style="font-size: 12pt\; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0) \; line-height: 1.5\;"><br></span><span style="font-size: 10pt\; font-fami ly: Arial\, Helvetica\, sans-serif\; color: rgb(64\, 64\, 64)\; font-weigh t: bold\;">Due to Security\, no walk-ins will be accepted. All attendees m ust register and log-in.<br></span><img style="vertical-align: bottom\; po sition: relative\; display: inline-block\; width: 50%\; min-width: 280px\; max-width: 280px\; background-image: none\; background-position-x: 0%\; b ackground-position-y: 0%\; background-size: auto\; background-repeat: repe at\; background-attachment: scroll\; background-origin: padding-box\; back ground-clip: border-box\; background-color: transparent\; height: auto\; m argin: 40px 0px 20px 0px\;" title="" src="https://enotice.vtools.ieee.org/ vtools_ui/media/display/d41fedc4-be33-462c-bd43-2f9daf14c95f" alt="" width ="226" height="84"><span style="font-size: 10pt\; font-family: Arial\, Hel vetica\, sans-serif\; color: rgb(64\, 64\, 64)\; font-weight: bold\;"><br> </span><span style="font-size: 10pt\; font-family: Arial\, Helvetica\, san s-serif\; color: rgb(0\, 0\, 0)\;">SPONSORED BY: <br>IEEE Computational In telligence Society Boston Chapter and the IEEE Boston AI Local Group</span ></div>\n</div>\n</div>\n</div>\n<div style="text-align: left\;"><span sty le="font-size: 12pt\; font-family: Arial\, Helvetica\, sans-serif\; color: rgb(0\, 0\, 0)\;"><br><br></span></div> END:VEVENT END:VCALENDAR