This research explores integration of a Large Language Model (LLM) fine-tuned to conversationally control the user interface (UI) for a Semantic Automation Layer (SAL). We condense SAL capabilities from prior work and prioritize with business analysts and data engineers via a Kano model, before implementing a prototypical UI. We augment the UI with our conversational engine and propose In-situ Prompt Engineering and learn from Human Feedback to smoothen the interaction and manipulation of UI through natural language commands. To evaluate the efficacy and usability of conversational control in various use-case scenarios, we conduct and report on an empirical interaction design user study. Our findings provide evidence supporting enhanced user engagement and satisfaction. We also observe significant increase of trust in AI after working with our conversational UI. This work generates areas for further refinement and research towards more intelligent, highly-integrated conversational UIs even beyond our application within Semantic Automation. We discuss our findings and point out next steps paving the way for future research and development in creating more intuitive and adaptive user interfaces.

We propose KnowGL, a tool that allows converting text into structured relational data represented as a set of ABox assertions compliant with the TBox of a given Knowledge Graph (KG), such as Wikidata. We address this problem as a sequence generation task by leveraging pre-trained sequence-to-sequence language models, e.g. BART. Given a sentence, we fine-tune such models to detect pairs of entity mentions and jointly generate a set of facts consisting of the full set of semantic annotations for a KG, such as entity labels, entity types, and their relationships. To showcase the capabilities of our tool, we build a web application consisting of a set of UI widgets that help users to navigate through the semantic data extracted from a given input text. We make the KnowGL model available at https://huggingface.co/ibm/knowgl-large.

For Ontology Alignment (OA), the task is to align semantically equivalent concepts and relations from different ontologies. This task plays a crucial role in many downstream tasks and applications in academia and industry. Since manually aligning ontologies is inefficient and costly, numerous approaches exist to do this automatically. However, most approaches are tailored to specific domains, are rule-based systems or based on feature engineering, and require external knowledge. The most recent advances in the field of OA rely on the widely proven effectiveness of pre-trained language models to represent the human-generated language that describes the entities in an ontology. However, these approaches additionally require sophisticated algorithms or Graph Neural Networks to exploit an ontology’s graphical structure to achieve state-of-the-art performance. In this work, we present NLFOA, or Natural Language Focused Ontology Alignment, which purely focuses on the natural language contained in ontologies to process the ontology’s semantics as well as graphical structure. An evaluation of our approach on common OA datasets shows superior results when finetuning with only a small number of training samples. Additionally, it demonstrates strong results in a zero-shot setting which could be employed in an active learning setup to reduce human labor when manually aligning ontologies significantly.

Michael Glass, Gaetano Rossiello, Md Faisal Mahbub Chowdhury, Ankita Naik, Pengshan Cai, Alfio Gliozzo. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

We present KaaPa (Knowledge Aware Answers from Pdf Analysis), an integrated solution for machine reading comprehension over both text and tables extracted from PDFs. KaaPa enables interactive question refinement using facets generated from an automatically induced Knowledge Graph. In addition it provides a concise summary of the supporting evidence for the provided answers by aggregating information across multiple sources. KaaPa can be applied consistently to any collection of documents in English with zero domain adaptation effort. We showcase the use of KaaPa for QA on scientific literature using the COVID-19 Open Research Dataset.

Saneem Chemmengath, Vishwajeet Kumar, Samarth Bharadwaj, Jaydeep Sen, Mustafa Canim, Soumen Chakrabarti, Alfio Gliozzo, Karthik Sankaranarayanan. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021.

Noun phrases and Relation phrases in open knowledge graphs are not canonicalized, leading to an explosion of redundant and ambiguous subject-relation-object triples.Existing approaches to solve this problem take a two-step approach.First, they generate embedding representations for both noun and relation phrases, then a clustering algorithm is used to group them using the embeddings as features.In this work, we propose Canonicalizing Using Variational Autoencoders (CUVA) 1 , a joint model to learn both embeddings and cluster assignments in an end-to-end approach, which leads to a better vector representation for the noun and relation phrases.Our evaluation over multiple benchmarks shows that CUVA outperforms the existing state-of-the-art approaches.Moreover, we introduce CANONICNELL, a novel dataset to evaluate entity canonicalization systems.

Automatically inducing high quality knowledge graphs from a given collection of documents still remains a challenging problem in AI.One way to make headway for this problem is through advancements in a related task known as slot filling.In this task, given an entity query in form of [ENTITY, SLOT, ?], a system is asked to 'fill' the slot by generating or extracting the missing value exploiting evidence extracted from relevant passage(s) in the given document collection.The recent works in the field try to solve this task in an end-to-end fashion using retrieval-based language models.In this paper, we present a novel approach to zero-shot slot filling that extends dense passage retrieval with hard negatives and robust training procedures for retrieval augmented generation models.Our model reports large improvements on both T-REx and zsRE slot filling datasets, improving both passage retrieval and slot value generation, and ranking at the top-1 position in the KILT leaderboard.Moreover, we demonstrate the robustness of our system showing its domain adaptation capability on a new variant of the TACRED dataset for slot filling, through a combination of zero/few-shot learning.We release the source code and pre-trained models 1 .

Michael Glass, Mustafa Canim, Alfio Gliozzo, Saneem Chemmengath, Vishwajeet Kumar, Rishav Chakravarti, Avi Sil, Feifei Pan, Samarth Bharadwaj, Nicolas Rodolfo Fauceglia. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Research on the task of Reading Comprehension style Question Answering (RCQA) has gained momentum in recent years due to the emergence of human annotated datasets and associated leaderboards, for example CoQA, HotpotQA, SQuAD, TriviaQA, etc. While state-of-the-art has advanced considerably, there is still ample opportunity to advance it further on some important variants of the RCQA task. In this paper, we propose a novel deep neural architecture, called TAP (Translucent Answer Prediction), to identify answers and evidence (in the form of supporting facts) in an RCQA task requiring multi-hop reasoning. TAP comprises two loosely coupled networks – Local and Global Interaction eXtractor (LoGIX) and Answer Predictor (AP). LoGIX predicts supporting facts, whereas AP consumes these predicted supporting facts to predict the answer span. The novel design of LoGIX is inspired by two key design desiderata – local context and global interaction– that we identified by analyzing examples of multi-hop RCQA task. The loose coupling between LoGIX and the AP reveals the set of sentences used by the AP in predicting an answer. Therefore, answer predictions of TAP can be interpreted in a translucent manner. TAP offers state-of-the-art performance on the HotpotQA (Yang et al. 2018) dataset – an apt dataset for multi-hop RCQA task – as it occupies Rank-1 on its leaderboard (https://hotpotqa.github.io/) at the time of submission.

This paper introduces Strict Partial Order Networks (SPON), a novel neural network architecture designed to enforce asymmetry and transitive properties as soft constraints. We apply it to induce hypernymy relations by training with is-a pairs. We also present an augmented variant of SPON that can generalize type information learned for in-vocabulary terms to previously unseen ones. An extensive evaluation over eleven benchmarks across different tasks shows that SPON consistently either outperforms or attains the state of the art on all but one of these benchmarks.

Michael Glass, Alfio Gliozzo, Rishav Chakravarti, Anthony Ferritto, Lin Pan, G P Shrivatsa Bhargav, Dinesh Garg, Avi Sil. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020.

Extracting lexico-semantic relations as graphstructured taxonomies, also known as taxonomy construction, has been beneficial in a variety of NLP applications.Recently Graph Neural Network (GNN) has shown to be powerful in successfully tackling many tasks.However, there has been no attempt to exploit GNN to create taxonomies.In this paper, we propose Graph2Taxo, a GNN-based cross-domain transfer framework for the taxonomy construction task.Our main contribution is to learn the latent features of taxonomy construction from existing domains to guide the structure learning of an unseen domain.We also propose a novel method of directed acyclic graph (DAG) generation for taxonomy construction.Specifically, our proposed Graph2Taxo uses a noisy graph constructed from automatically extracted noisy hyponym-hypernym candidate pairs, and a set of taxonomies for some known domains for training.The learned model is then used to generate taxonomy for a new unknown domain given a set of terms for that domain.Experiments on benchmark datasets from science and environment domains show that our approach attains significant improvements correspondingly over the state of the art.

We propose a novel approach to learn representations of relations expressed by their textual mentions. In our assumption, if two pairs of entities belong to the same relation, then those two pairs are analogous. We collect a large set of analogous pairs by matching triples in knowledge bases with web-scale corpora through distant supervision. This dataset is adopted to train a hierarchical siamese network in order to learn entity-entity embeddings which encode relational information through the different linguistic paraphrasing expressing the same relation. The model can be used to generate pre-trained embeddings which provide a valuable signal when integrated into an existing neural-based model by outperforming the state-of-the-art methods on a relation extraction task.

Analogy is a fundamental component of the way we think and process thought. Solving a word analogy problem, such as mason is to stone as carpenter is to wood, requires capabilities in recognizing the implicit relations between the two word pairs. In this paper, we describe the analogy problem from a computational linguistics point of view and explore its use to address relation extraction tasks. We extend a relational model that has been shown to be effective in solving word analogies and adapt it to the relation extraction problem. Our experiments show that this approach outperforms the state-of-the-art methods on a relation extraction dataset, opening up a new research direction in discovering implicit relations in text through analogical reasoning.

State-of-the-art relation extraction ap-proaches are only able to recognize relationships between mentions of entity arguments stated explicitly in the text and typically localized to the same sentence.However, the vast majority of relations are either implicit or not sententially localized.This is a major problem for Knowledge Base Population, severely limiting recall.In this paper we propose a new methodology to identify relations between two entities, consisting of detecting a very large number of unary relations, and using them to infer missing entities.We describe a deep learning architecture able to learn thousands of such relations very efficiently by using a common deep learning based representation.Our approach largely outperforms state of the art relation extraction technology on a newly introduced web scale knowledge base population benchmark, that we release to the research community.

Previous studies have highlighted the necessity for entity linking systems to capture the local entity-mention similarities and the global topical coherence. We introduce a novel framework based on convolutional neural networks and recurrent neural networks to simultaneously model the local and global features for entity linking. The proposed model benefits from the capacity of convolutional neural networks to induce the underlying representations for local contexts and the advantage of recurrent neural networks to adaptively compress variable length sequences of predictions for global constraints. Our evaluation on multiple datasets demonstrates the effectiveness of the model and yields the state-of-the-art performance on such datasets. In addition, we examine the entity linking systems on the domain adaptation setting that further demonstrates the cross-domain robustness of the proposed model.

In this paper we examine the lexical substitu-tion task for the medical domain. We adapt the current best system from the open domain, which trains a single classifier for all instances using delexicalized features. We show sig-nificant improvements over a strong baseline coming from a distributional thesaurus (DT). Whereas in the open domain system, features derived from WordNet show only slight im-provements, we show that its counterpart for the medical domain (UMLS) shows a signif-icant additional benefit when used for feature generation. 1

Many forms of word relatedness have been developed, providing different perspec-tives on word similarity. We introduce a Bayesian probabilistic tensor factoriza-tion model for synthesizing a single word vector representation and per-perspective linear transformations from any number of word similarity matrices. The result-ing word vectors, when combined with the per-perspective linear transformation, ap-proximately recreate while also regulariz-ing and generalizing, each word similarity perspective. Our method can combine manually cre-ated semantic resources with neural word embeddings to separate synonyms and antonyms, and is capable of generaliz-ing to words outside the vocabulary of any particular perspective. We evaluated the word embeddings with GRE antonym questions, the result achieves the state-of-the-art performance. 1

Automatic open-domain Question Answering has been a long standing research challenge in the AI community. IBM Research undertook this challenge with the design of the DeepQA architecture and the implementation of Watson. This paper addresses a specific subtask of Deep QA, consisting of predicting the Lexical Answer Type (LAT) of a question. Our approach is completely unsupervised and is based on PRISMATIC, a large-scale lexical knowledge base automatically extracted from a Web corpus. Experiments on the Jeopardy! data shows that it is possible to correctly predict the LAT in a substantial number of questions. This approach can be used for general purpose knowledge acquisition tasks such as frame induction from text.

We present a semi-supervised technique for word sense disambiguation that exploits external knowledge acquired in an unsupervised manner. In particular, we use a combination of basic kernel functions to independently estimate syntagmatic and domain similarity, building a set of word-expert classifiers that share a common domain model acquired from a large corpus of unlabeled data. The results show that the proposed approach achieves state-of-the-art performance on a wide range of lexical sample tasks and on the English all-words task of Senseval-3, although it uses a considerably smaller number of training examples than other methods.

This paper investigates conceptually and empirically the novel sense matching task, which requires to recognize whether the senses of two synonymous words match in context. We suggest direct approaches to the problem, which avoid the intermediate step of explicit word sense disambiguation, and demonstrate their appealing advantages and stimulating potential for future research.

Cross-language Text Categorization is the task of assigning semantic classes to documents written in a target language (e.g. English) while the system is trained using labeled documents in a source language (e.g. Italian).In this work we present many solutions according to the availability of bilingual resources, and we show that it is possible to deal with the problem even when no such resources are accessible. The core technique relies on the automatic acquisition of Multilingual Domain Models from comparable corpora.Experiments show the effectiveness of our approach, providing a low cost solution for the Cross Language Text Categorization task. In particular, when bilingual dictionaries are available the performance of the categorization gets close to that of monolingual text categorization.

In this paper we present a supervised Word Sense Disambiguation methodology, that exploits kernel methods to model sense distinctions. In particular a combination of kernel functions is adopted to estimate independently both syntagmatic and domain similarity. We defined a kernel function, namely the Domain Kernel, that allowed us to plug "external knowledge" into the supervised learning process. External knowledge is acquired from unlabeled data in a totally unsupervised way, and it is represented by means of Domain Models. We evaluated our methodology on several lexical sample tasks in different languages, outperforming significantly the state-of-the-art for each of them, while reducing the amount of labeled training data required for learning.

This paper explores the role of domain information in word sense disambiguation. The underlying hypothesis is that domain labels, such as MEDICINE, ARCHITECTURE and SPORT, provide a useful way to establish semantic relations among word senses, which can be profitably used during the disambiguation process. Results obtained at the SENSEVAL-2 initiative confirm that for a significant subset of words domain information can be used to disambiguate with a very high level of precision.