More Like this

1. Graph Enhanced BERT for Query Understanding

   Li, Juanhui ; Zeng, Wei ; Cheng, Suqi ; Ma, Yao ; Tang, Jiliang ; Wang, Shuaiqiang ; Yin, Dawei ( July 2023 , International ACM SIGIR Conference on Research and Development in Information Retrieval)

   Query understanding plays a key role in exploring users’ search intents. However, it is inherently challenging since it needs to capture semantic information from short and ambiguous queries and often requires massive task-specific labeled data. In recent years, pre-trained language models (PLMs) have advanced various natural language processing tasks because they can extract general semantic information from large-scale corpora. However, directly applying them to query understanding is sub-optimal because existing strategies rarely consider to boost the search performance. On the other hand, search logs contain user clicks between queries and urls that provide rich users’ search behavioral information on queries beyond their content. Therefore, in this paper, we aim to fill this gap by exploring search logs. In particular, we propose a novel graph-enhanced pre-training framework, GE-BERT, which leverages both query content and the query graph to capture both semantic information and users’ search behavioral information of queries. Extensive experiments on offline and online tasks have demonstrated the effectiveness of the proposed framework. 

   more » « less
2. The worst graph layout algorithm ever

   Di Bartolomeo, Sara ; Lang, Matěj ; Dunne, Cody ( January 2022 , Proc. alt.VIS workshop at IEEE VIS)

   Graph layout algorithms strive to improve the utility of node-link visualizations or graph drawings by optimizing for readability criteria. One such criteria that has been widely used is to count edge crossings. Prior work has focused solely on minimizing the number of edge crossings, including provably-optimal layout algorithms for layered graphs. The research community has completely ignored the other side of the coin — can we optimally maximize edge crossings? This paper answers this question in the affirmative. Our WORSTisfimal layout algorithm produces the most unreadable layered graph drawing. It does so by using linear programming to produce a provably-optimally-awful solution. We hope that this groundbreaking result opens up an entirely new field of inquiry for graph drawing researchers — optimally-worst layout algorithms. 

   more » « less
3. NanoMine: A Knowledge Graph for Nanocomposite Materials Science, Lecture Notes in Computer Science, vol 12507, doi.org/10.1007/978-3-030-62466-8_10

   https://doi.org/10.1007/978-3-030-62466-8_10  

   McCusker J.P., Keshan N. ( October 2020 , International Semantic Web Conference, ISBN: 978-3-030-62466-8)

   null (Ed.)

   Knowledge graphs can be used to help scientists integrate and explore their data in novel ways. NanoMine, built with the Whyis knowledge graph framework, integrates diverse data from over 1,700 polymer nanocomposite experiments. Polymer nanocomposites (polymer materials with nanometer-scale particles embedded in them) exhibit complex changes in their properties depending upon their composition or processing methods. Building an overall theory of how nanoparticles interact with the polymer they are embedded in therefore typically has to rely on an integrated view across hundreds of datasets. Because the NanoMine knowledge graph is able to integrate across many experiments, materials scientists can explore custom visualizations and, with minimal semantic training, produce custom visualizations of their own. NanoMine provides access to experimental results and their provenance in a linked data format that conforms to well-used semantic web ontologies and vocabularies (PROV-O, Schema.org, and the Semanticscience Integrated Ontology). We curated data described by an XML schema into an extensible knowledge graph format that enables users to more easily browse, filter, and visualize nanocomposite materials data. We evaluated NanoMine on the ability for material scientists to produce visualizations that help them explore and understand nanomaterials and assess the diversity of the integrated data. Additionally, NanoMine has been used by the materials science community to produce an integrated view of a journal special issue focusing on data sharing, demonstrating the advantages of sharing data in an interoperable manner. 

   more » « less
4. Building trustworthy NeuroSymbolic AI Systems: Consistency, reliability, explainability, and safety

   https://doi.org/10.1002/aaai.12149  

   Gaur, Manas ; Sheth, Amit ( February 2024 , AI Magazine)

   Explainability and Safety engender trust. These require a model to exhibit consistency and reliability. To achieve these, it is necessary to use and analyzedataandknowledgewith statistical and symbolic AI methods relevant to the AI application––neither alone will do. Consequently, we argue and seek to demonstrate that the NeuroSymbolic AI approach is better suited for making AI a trusted AI system. We present the CREST framework that shows howConsistency,Reliability, user‐levelExplainability, andSafety are built on NeuroSymbolic methods that use data and knowledge to support requirements for critical applications such as health and well‐being. This article focuses on Large Language Models (LLMs) as the chosen AI system within the CREST framework. LLMs have garnered substantial attention from researchers due to their versatility in handling a broad array of natural language processing (NLP) scenarios. As examples, ChatGPT and Google's MedPaLM have emerged as highly promising platforms for providing information in general and health‐related queries, respectively. Nevertheless, these models remain black boxes despite incorporating human feedback and instruction‐guided tuning. For instance, ChatGPT can generateunsafe responsesdespite instituting safety guardrails. CREST presents a plausible approach harnessing procedural and graph‐based knowledge within a NeuroSymbolic framework to shed light on the challenges associated with LLMs.

    

   more » « less
5. Visual Query Answering by Entity-Attribute Graph Matching and Reasoning

   Xiong, Peixi ; Zhan, Huayi ; Wang, Xin ; Shinha, Baivab ; Wu, Ying ( June 2019 , Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition)

   Visual Query Answering (VQA) is of great significance in offering people convenience: one can raise a question for details of objects, or high-level understanding about the scene, over an image. This paper proposes a novel method to address the VQA problem. In contrast to prior works, our method that targets single scene VQA, replies on graph-based techniques and involves reasoning. In a nutshell, our approach is centered on three graphs. The first graph, referred to as inference graph GI , is constructed via learning over labeled data. The other two graphs, referred to as query graph Q and entity-attribute graph GEA, are generated from natural language query Qnl and image Img, that are issued from users, respectively. As GEA often does not take sufficient information to answer Q, we develop techniques to infer missing information of GEA with GI . Based on GEA and Q, we provide techniques to find matches of Q in GEA, as the answer of Qnl in Img. Unlike commonly used VQA methods that are based on end-to-end neural networks, our graph-based method shows well-designed reasoning capability, and thus is highly interpretable. We also create a dataset on soccer match (Soccer-VQA) with rich annotations. The experimental results show that our approach outperforms the state-of-the-art method and has high potential for future investigation. 

   more » « less