Kuldeep Singh

We introduce BioLinkerAI, a neuro-symbolic framework for biomedical entity linking that integrates symbolic (domain-specific and linguistic rules) and sub-symbolic (large language models) components. Unlike traditional approaches requiring extensive labeled training data, BioLinkerAI harnesses a knowledge base and rules for candidate generation, while a pre-trained LLM handles final disambiguation. This combination ensures adaptability to diverse biomedical knowledge bases and complex entity mentions. Empirical evaluations show that BioLinkerAI surpasses state-of-the-art benchmarks, notably increasing unseen data accuracy from 65.4 to 78.5 without relying on extensive labeled datasets.

Question answering (QA) over knowledge graphs (KGs) is an essential task that maps a user’s utterance to a query over a KG to retrieve the correct answer. Earlier methods in this field relied heavily on predefined templates and rules, which had limited adaptability and learning capability. Recent research has made significant strides in answering straightforward questions, and there has been notable success in tackling more intricate queries as well. However, a key challenge remains that, often, a single round of question and answer is not enough. Users might have follow-up questions that delve deeper into a subject, relying on information from their initial queries. This situation is particularly common in conversational settings, where each new question might refer back to earlier topics or answers. In this chapter, we explore advanced techniques for handling such conversational QA over knowledge graphs. We leverage deep neural networks and multi-task learning approaches to create systems that can understand and respond to a series of interconnected questions. By focusing on these conversational aspects and the nuances between different types of queries, we aim to bridge a significant gap in current research, offering more dynamic and context-aware systems that can adapt to the evolving nature of human inquiry.

We present the Evolving Graph Fourier Transform (EFT), the first invertible spectral transform that captures evolving representations on temporal graphs. We motivate our work by the inadequacy of existing methods for capturing the evolving graph spectra, which are also computationally expensive due to the temporal aspect along with the graph vertex domain. We view the problem as an optimization over the Laplacian of the continuous time dynamic graph. Additionally, we propose pseudo-spectrum relaxations that decompose the transformation process, making it highly computationally efficient. The EFT method adeptly captures the evolving graph's structural and positional properties, making it effective for downstream tasks on evolving graphs. Hence, as a reference implementation, we develop a simple neural model induced with EFT for capturing evolving graph spectra. We empirically validate our theoretical findings on a number of large-scale and standard temporal graph benchmarks and demonstrate that our model achieves state-of-the-art performance.

Document-level relation extraction (DocRE) poses the challenge of identifying relationships between entities within a document. Existing approaches rely on logical reasoning or contextual cues from entities. This paper reframes document-level RE as link prediction over a Knowledge Graph (KG) with distinct benefits: 1) Our approach amalgamates entity context and document-derived logical reasoning, enhancing link prediction quality. 2) Predicted links between entities offer interpretability, elucidating employed reasoning. We evaluate our approach on benchmark datasets - DocRED, ReDocRED, and DWIE. The results indicate that our proposed method outperforms the state-of-the-art models and suggests that incorporating context-based Knowledge Graph link prediction techniques can enhance the performance of document-level relation extraction models.

Document-level relation extraction (DocRE) poses the challenge of identifying relationships between entities within a document as opposed to the traditional RE setting where a single sentence is input. Existing approaches rely on logical reasoning or contextual cues from entities. This paper reframes document-level RE as link prediction over a knowledge graph with distinct benefits: 1) Our approach combines entity context with document-derived logical reasoning, enhancing link prediction quality. 2) Predicted links between entities offer interpretability, elucidating employed reasoning. We evaluate our approach on three benchmark datasets: DocRED, ReDocRED, and DWIE. The results indicate that our proposed method outperforms the state-of-the-art models and suggests that incorporating context-based link prediction techniques can enhance the performance of document-level relation extraction models.

Monika Jain, Raghava Mutharaju, Kuldeep Singh, Ramakanth Kavuluru. Findings of the Association for Computational Linguistics ACL 2024. 2024.

Learning on evolving(dynamic) graphs has caught the attention of researchers as static methods exhibit limited performance in this setting. The existing methods for dynamic graphs learn spatial features by local neighborhood aggregation, which essentially only captures the low pass signals and local interactions. In this work, we go beyond current approaches to incorporate global features for effectively learning representations of a dynamically evolving graph. We propose to do so by capturing the spectrum of the dynamic graph. Since static methods to learn the graph spectrum would not consider the history of the evolution of the spectrum as the graph evolves with time, we propose an approach to learn the graph wavelets to capture this evolving spectra. Further, we propose a framework that integrates the dynamically captured spectra in the form of these learnable wavelets into spatial features for incorporating local and global interactions. Experiments on eight standard datasets show that our method significantly outperforms related methods on various tasks for dynamic graphs.