GraphNews

In my previous post,

Reasoning with Knowledge Graph Clustering in Retrieval-Augmented Generation Systems 🔲 ⚫ Retrieval-augmented generation (RAG) systems have gained immense… | 35 comments on LinkedIn

The secret of AI: it’s all about building your Knowledge Graphs. This open secret deserves more spotlight: a LLM, at its core, is fundamentally a database, a… | 69 comments on LinkedIn

Injecting KGs in RAG in pre-processing, post-processing, chunk extraction, document and contextual hierarchies, with a concrete example.

Python open source projects; natural language meets graph technologies; graph topological transformations; graph levels of detail (abstraction layers)

Use Gartner’s impact radar for generative AI to plan investments and strategy with four key themes in mind: ☑️Model-related innovations ☑️Model performance and AI safety ☑️Model build and data-related ☑️AI-enabled applications Explore all 25 technologies and trends: https://www.gartner.com/en/articles/understand-and-exploit-gen-ai-with-gartner-s-new-impact-radar

What do we mean when we say something is a kind of thing? I’ve been wrestling with that question a great deal of late, partly because I think the role of the ontologist transcends the application of knowledge graphs, especially as I’ve watched LLMs and Llamas become a bigger part of the discussion.

This blog post summarizes the findings of our new contribution:

Neural algorithmic reasoning focuses on building models that can execute classic algorithms. It allows one to combine the advantages of neural networks, such as handling raw and noisy input data, with theoretical guarantees and strong generalization of algorithms. Assuming we have a neural network capable of solving a classic algorithmic task, we can incorporate it into a more complex pipeline and train end-to-end. For instance, if we have a neural solver aligned to the shortest path problem, it can be used as a building block for a routing system that accounts for complex and dynamically changing traffic conditions. In our work [ref1], we study algorithmic reasoners trained only from input-output pairs, in contrast to current state-of-the-art approaches that utilize the trajectory of a given algorithm. We propose several architectural modifications and demonstrate how standard contrastive learning techniques can regularize intermediate computations of the models without appealing to any predefined algorithm’s trajectory.

Fusion Knowledge Graphs and Language Models Through Compatible Generative Modeling 🌙 Knowledge graphs (KGs) and large language models (LLMs) have…

Knowledge engineering is a discipline that focuses on the creation and maintenance of processes that generate and apply knowledge. Traditionally, knowledge engineering approaches have focused on knowledge expressed in formal languages. The emergence of large language models and their capabilities to effectively work with natural language, in its broadest sense, raises questions about the foundations and practice of knowledge engineering. Here, we outline the potential role of LLMs in knowledge engineering, identifying two central directions: 1) creating hybrid neuro-symbolic knowledge systems; and 2) enabling knowledge engineering in natural language. Additionally, we formulate key open research questions to tackle these directions.

Orchestrating Efficient Reasoning Over Knowledge Graphs with LLM Compiler Frameworks 🔵 (published in Towards Data Science) Recent innovations in Large…

Optimizing Retrieval-Augmented Generation (RAG) by Selective Knowledge Graph Conditioning 🔝 (published in Towards Data Science) Generative pre-trained… | 15 comments on LinkedIn

A broad and deep body of on-going research – hundreds of experiments! – has shown quite conclusively that knowledge graphs are essential to guide, complement, and enrich LLMs in systematic ways. The very wide variety of tests over domains and possible combinations of KGs and LLMs attests to the robu

Hallucination is a common problem when working with large language models (LLMs). LLMs generate fluent and coherent text but often generate…

Knowledge graph embedding models (KGEMs) developed for link prediction learn vector representations for graph entities, known as embeddings. A common tacit assumption is the KGE entity similarity assumption, which states that these KGEMs retain the graph's structure within their embedding space, i.e., position similar entities close to one another. This desirable property make KGEMs widely used in downstream tasks such as recommender systems or drug repurposing. Yet, the alignment of graph similarity with embedding space similarity has rarely been formally evaluated. Typically, KGEMs are assessed based on their sole link prediction capabilities, using ranked-based metrics such as Hits@K or Mean Rank. This paper challenges the prevailing assumption that entity similarity in the graph is inherently mirrored in the embedding space. Therefore, we conduct extensive experiments to measure the capability of KGEMs to cluster similar entities together, and investigate the nature of the underlying factors. Moreover, we study if different KGEMs expose a different notion of similarity. Datasets, pre-trained embeddings and code are available at: https://github.com/nicolas-hbt/similar-embeddings.

The numerical representation of data is often ambiguous. This leads to a gauge theoretic view on data, requiring covariant or equivariant neural networks which are reviewed in this blog post.

Knowledge Graph Embeddings as a Bridge between Symbolic and Subsymbolic AI 🌉 The Resurgence of Structure The pendulum in AI is swinging back from purely…

Our manuscript on using Large Language Models and Retrieval Augmented Generation for creating ontology terms is up on arXiv! https://lnkd.in/d62JPtiH, lead…

Learn how to build knowledge graphs using Python and large language models (LLMs) to create intricate interconnected knowledge maps

Augmenting Large Language Models with Hybrid Knowledge Architectures 🧐 Vector search or Knowledge graph ? Why not both at the same time… Retrieval…

By request, here are the slides from our #neurips2023 presentation yesterday! We really enjoyed the opportunity to present the different aspects of the work… | 18 comments on LinkedIn

Here's the video for my talk @ K1st World Symposium 2023 about the intersections of KGs and LLMs: https://lnkd.in/gugB8Yjj and also the slides, plus related…

While LLMs have shown remarkable prowess in understanding and generating text, they have a critical limitation.

The evidence for the massive impact of KGs in NLQ keeps piling up - Here's one more paper that shows that knowledge graph based RAG (retrieval-augmentation)…

An architecture review of Neo4j as both vector and graph store

On August 14, 2023, the paper Natural Language is All a Graph Needs by Ruosong Ye, Caiqi Zhang, Runhui Wang, Shuyuan Xu and Yongfeng Zhang…

This week Alexander Erdl and I did our last #GoingMeta of the year, and we talked about Advanced #RAG patterns with #KnowledgeGraphs and vector search. We…