I used to be at KGC (Information Graph Convention) 2024, which is going on Might 6-10 at Cornell Tech. I used to be presenting (just about) at their Well being Care and Life Sciences (HCLS) workshop, so my audio system go was solely legitimate for at present for the HCLS portion of KGC. My journey report covers a number of talks that I attended right here. Attending just about was a bit chaotic as periods went over typically, so that you would possibly go away a session to attend one other, solely to search out that it hadn’t began but. That is laborious to forsee, now we have confronted this problem ourselves the primary time we moved an inside convention from in-person to hybrid.
KGs in RAG (Tom Smoker, WhatWhyHow.AI)
I’ve been working with Giant Language Fashions (LLMs) and Retrieval Augmented Technology (RAG) for nearly a 12 months now, and I went to this discuss hoping for insights on the right way to use graphs as enter to RAG techniques. Understandably, the speaker spent a while masking the fundamentals, which I personally didn’t discover very fruitful. Nonetheless, there have been some nuggets of knowledge I received out of the discuss. First, the RAG pipelines can decrease the danger of hallucinations by utilizing LLMs for planning and reasoning, however with out delegating to LLMs for factual info. And second, an agent structure can extra effectively use smaller sub-graphs which may usually be generated dynamically in Closed World fashions.
A facet dialogue on chat additionally yielded a paper reference Getting from Generative AI to Reliable AI: what LLMs could study from Cyc (Lenat and Marcus, 2023). The paper appears to be like actually fascinating on an preliminary skim and I plan to learn in additional element later.
Information Graphs for Precision Oncology (Krishna Bulusu, AstraZeneca)
A pleasant overview of purposes of Information Graph (KG) to Drug Discovery (DD). DD makes an attempt to use KG to resolve 3 most important issues: (1) discover gene inflicting illness (2) match drug with illness and (3) (drug, gene, illness) as a elementary relationship in DD. The speaker identified that the large benefit of KGs is Explainability. He additionally talked about using graph clustering for node stratification.
Combining graph and vector illustration for environment friendly info retrieval (Peio Popov, Ontotext)
This was a presentation from OntoText the place they demonstrated new options constructed into their GraphDB database. This was of curiosity to me personally since our KG can be constructed utilizing GraphDB. Particularly they’ve built-in LLM and vector search help into their merchandise to allow them to be invoked from a SPARQL question. This offers GraphDB customers the ability to mix these strategies in the identical name relatively than construct multi-stage pipelines.
I additionally realized the excellence between Semantic, Full textual content and Vector Search as ones primarily based off KG, Lucene (or Lucene-like) indexes and vector search platforms, I’d beforehand conflate the primary and third.
Information Engineering in Medical Choice Assist: When a Graph Representational Mannequin will not be sufficient (Maulik Kamdar, Optum)
This was a presentation from my ex-colleague Maulik Kamdar. He talks about challenges in Medical Choice Assist (CDS) the place a KG alone is inadequate. Particularly the case he’s contemplating the place a number of third social gathering ontologies should be aligned into one KG. On this scenario, related ideas are mixed into ValueSets, that are then composed with naked ideas or with one another to kind Medical Guidelines. Medical Guidelines are additional mixed to kind Medical Calculators or Questionnaires, that are then mixed to kind Choice Timber and Flowcharts, that are then mixed into Medical Tips. I’m in all probability biased given our frequent historical past, however I discovered this discuss to be essentially the most academic for me.
Information Graphs, Theorem Provers and Language Fashions (Vijay Saraswat and Nikolaos Vasiloglou)
The audio system mentioned the function of self-discovery, In-Context Studying (ICL), symbiotic integration of KG with search, and Graph RAG in reasoning engines powered by KG and LLM. They characterize an Agent as an LLM primarily based black field that is supplied with pairs of input-output situations to study some unknown operate (much like ML fashions). They describe ICL as studying via few shot and plenty of shot examples. In addition they speak about utilizing the output of KG to fact-check / improve LLMs and utilizing LLMs to generate assertions that can be utilized to create a KG. Their demo exhibits how an LLM is ready to study to generate a Datalog like graph question language from textual content prompts utilizing few-shot examples.
The speaker made reference to the next three papers in help of the strategies he was describing, which I’ve duly added to my studying checklist.
A Scalable and Sturdy Named Entity Recognition and Linking System for a Medical Healthcare Information Graph (Sujit Pal, Elsevier Well being)
This was my discuss. I had initially supposed to attend in particular person nevertheless it appeared wasteful to fly throughout the nation to ship a 5-minute presentation. It did take a little bit of planning to current remotely however I realized two helpful life classes.
- You may generate a presentation video from MS Powerpoint. Merely create your slides and file a slideshow the place you file your self narrating your presentation. As soon as completed, export as an MP4 and add to Youtube or different video service.
- You may print posters on-line and have them delivered to another person.
Big due to my colleague Tom Woodcock who attended in particular person, and who was sort sufficient to hold and cling my poster on the convention for me, and who additionally agreed to current my slideshow for me (though I feel that ultimately he didn’t must). Many thanks additionally to my ex-colleague Helena Deus (a part of the HCLS organizing workforce), who helped stroll me via to a workable resolution and was instrumental in my discuss being delivered efficiently. Additionally due to Leah Walton from the HCLS organizing workforce, for supporting me in my try to current remotely.
Right here is the Youtube video for my 5-minute presentation in case you have an interest. It’s a bit high-level since I had solely 5 minutes to cowl every part, however there is a bit more info within the poster under.
Graphs for good – Speculation era for Uncommon Illness Therapy (Brian Martin, AbbVie)
This presentation revolves round a graph that connects illnesses to medicine through illness variants, gene, pathway, gene and compound entities. This was used to discover a remedy for a uncommon illness utilizing present medicines. It was later prolonged to search out candidate cures for a gaggle of 20 most uncared for illnesses worldwide. The audio system verified that outcomes for Dengue fever correlates effectively with beforehand identified info, thus supporting the veracity of the strategy. The paper describing this work is Leveraging a Billion-Edge Information Graph for Drug Re-purposing and Goal Prioritization utilizing Genomically-Knowledgeable Subgraphs (Martin et al, 2022).
Producing and Querying Graphs with LLM (Brian Martin, Subha Madhavan, Berenice Wulbrecht)
Panel dialogue the place numerous methods for producing and querying graphs utilizing LLMs have been mentioned. Entertaining (and considerably predictable) comparisons of Property Graphs vs RDF graphs to Ford and Ferrari vehicles, and the way LLMs rework them into Teslas (with its self-driving expertise). In addition they speak about extracting assertions from a corpus of paperwork to create a KG custom-made for the corpus, after which utilizing the KG to fact-check the output of the LLM for RAG queries towards that corpus.
Total, I feel it was a fantastic convention. Realized rather a lot, would love to return and current right here sooner or later, hopefully this time in particular person.
