BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Chicago
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:CDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T010000
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:CST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260520T181339Z
UID:D9595BC9-38B3-4868-9EB9-4773E6704609
DTSTART;TZID=America/Chicago:20260520T120000
DTEND;TZID=America/Chicago:20260520T130000
DESCRIPTION:Retrieval-augmented generation (RAG) has become a widely adopte
 d technique for reducing hallucinations and injecting domain knowledge int
 o large language models. Most implementations rely on vector similarity se
 arch. This is effective for semantic matching but limited when accurate re
 sponses require understanding relationships between entities\, not just pr
 oximity in embedding space.\n\nThis session introduces knowledge graphs as
  a complementary grounding architecture for LLM pipelines. We&#39;ll examine w
 here vector RAG breaks down in scenarios like:\n\n● Multi-hop reasoning\
 n\n● Entity disambiguation\n\n● Relational context\n\nThe session will
  demonstrate how a property graph model captures what embeddings obscure. 
 Attendees will see how GraphRAG pipelines combine the recall strengths of 
 vector search with the precision of graph traversal\, producing responses 
 that are more accurate and inherently more explainable.\n\nNo prior graph 
 database experience is required\, as we’ll be introducing the fundamenta
 l building blocks of the labeled property graph paradigm. Engineers famili
 ar with RAG pipelines and transformer-based models will find the concepts 
 immediately applicable.\n\nSpeaker(s): Rob Martin\, \n\nVirtual: https://e
 vents.vtools.ieee.org/m/557613
LOCATION:Virtual: https://events.vtools.ieee.org/m/557613
ORGANIZER:bishopm@acm.org
SEQUENCE:30
SUMMARY: Intro to Knowledge Graphs and Grounding LLMs w/ Knowledge Graphs\,
  including a few Neo4j real-world examples
URL;VALUE=URI:https://events.vtools.ieee.org/m/557613
X-ALT-DESC:Description: &lt;br /&gt;&lt;p class=&quot;MsoNormal&quot; style=&quot;margin-left: 30.0
 pt\;&quot;&gt;&lt;span lang=&quot;EN&quot;&gt;Retrieval-augmented generation (RAG) has become a wi
 dely adopted technique for reducing hallucinations and injecting domain kn
 owledge into large language models. Most implementations rely on vector si
 milarity search. This is effective for semantic matching but limited when 
 accurate responses require understanding&amp;nbsp\;&lt;em&gt;relationships&lt;/em&gt; betw
 een entities\, not just proximity in embedding space.&lt;/span&gt;&lt;/p&gt;\n&lt;p class
 =&quot;MsoNormal&quot; style=&quot;margin-left: 30.0pt\;&quot;&gt;&lt;span lang=&quot;EN&quot;&gt;This session in
 troduces knowledge graphs as a complementary grounding architecture for LL
 M pipelines. We&#39;ll examine where vector RAG breaks down in scenarios like:
 &lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;text-indent: -.25in\; mso-list: l
 0 level1 lfo1\; margin: 12.0pt 0in .0001pt 66.0pt\;&quot;&gt;&lt;!-- [if !supportList
 s]--&gt;&lt;span lang=&quot;EN&quot;&gt;&lt;span style=&quot;mso-list: Ignore\;&quot;&gt;●&lt;span style=&quot;font
 : 7.0pt &#39;Times New Roman&#39;\;&quot;&gt;&amp;nbsp\;&amp;nbsp\;&amp;nbsp\;&amp;nbsp\; &lt;/span&gt;&lt;/span&gt;&lt;/
 span&gt;&lt;!--[endif]--&gt;&lt;span lang=&quot;EN&quot;&gt;Multi-hop reasoning&lt;/span&gt;&lt;/p&gt;\n&lt;p clas
 s=&quot;MsoNormal&quot; style=&quot;margin-left: 66.0pt\; text-indent: -.25in\; mso-list:
  l0 level1 lfo1\;&quot;&gt;&lt;!-- [if !supportLists]--&gt;&lt;span lang=&quot;EN&quot;&gt;&lt;span style=&quot;
 mso-list: Ignore\;&quot;&gt;●&lt;span style=&quot;font: 7.0pt &#39;Times New Roman&#39;\;&quot;&gt;&amp;nbsp
 \;&amp;nbsp\;&amp;nbsp\;&amp;nbsp\; &lt;/span&gt;&lt;/span&gt;&lt;/span&gt;&lt;!--[endif]--&gt;&lt;span lang=&quot;EN&quot;
 &gt;Entity disambiguation&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;text-indent
 : -.25in\; mso-list: l0 level1 lfo1\; margin: 0in 0in 12.0pt 66.0pt\;&quot;&gt;&lt;!-
 - [if !supportLists]--&gt;&lt;span lang=&quot;EN&quot;&gt;&lt;span style=&quot;mso-list: Ignore\;&quot;&gt;
 ●&lt;span style=&quot;font: 7.0pt &#39;Times New Roman&#39;\;&quot;&gt;&amp;nbsp\;&amp;nbsp\;&amp;nbsp\;&amp;nbs
 p\; &lt;/span&gt;&lt;/span&gt;&lt;/span&gt;&lt;!--[endif]--&gt;&lt;span lang=&quot;EN&quot;&gt;Relational context&lt;
 /span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;margin-left: 30.0pt\;&quot;&gt;&lt;span lang=
 &quot;EN&quot;&gt;The session will demonstrate how a property graph model captures what
  embeddings obscure. Attendees will see how GraphRAG pipelines combine the
  recall strengths of vector search with the precision of graph traversal\,
  producing responses that are more accurate and inherently more explainabl
 e.&lt;/span&gt;&lt;span lang=&quot;EN&quot;&gt;&amp;nbsp\;&lt;/span&gt;&lt;/p&gt;\n&lt;p class=&quot;MsoNormal&quot; style=&quot;m
 argin-left: 30.0pt\;&quot;&gt;&lt;span lang=&quot;EN&quot;&gt;No prior graph database experience i
 s required\, as we&amp;rsquo\;ll be introducing the fundamental building block
 s of the labeled property graph paradigm. Engineers familiar with RAG pipe
 lines and transformer-based models will find the concepts immediately appl
 icable.&lt;/span&gt;&lt;/p&gt;
END:VEVENT
END:VCALENDAR