Alfio Gliozzo

Michael Glass, Gaetano Rossiello, Md Faisal Mahbub Chowdhury et al. · ibm-research

As demonstrated by GPT-3 and T5, transformers grow in capability as parameter spaces become larger and larger. However, for tasks that require a large amount of knowledge, non-parametric memory allows models to grow dramatically with a sub-linear increase in computational cost and GPU memory requirements. Recent models such as RAG and REALM have introduced retrieval into conditional generation. These models incorporate neural initial retrieval from a corpus of passages. We build on this line of research, proposing Re2G, which combines both neural initial retrieval and reranking into a BART-based sequence-to-sequence generation. Our reranking approach also permits merging retrieval results from sources with incomparable scores, enabling an ensemble of BM25 and neural initial retrieval. To train our system end-to-end, we introduce a novel variation of knowledge distillation to train the initial retrieval, reranker, and generation using only ground truth on the target sequence output. We find large gains in four diverse tasks: zero-shot slot filling, question answering, fact-checking, and dialog, with relative gains of 9% to 34% over the previous state-of-the-art on the KILT leaderboard. We make our code available as open source at https://github.com/IBM/kgi-slot-filling/tree/re2g.

Md Faisal Mahbub Chowdhury, Michael Glass, Gaetano Rossiello et al. · ibm-research

In this paper, we present a system to showcase the capabilities of the latest state-of-the-art retrieval augmented generation models trained on knowledge-intensive language tasks, such as slot filling, open domain question answering, dialogue, and fact-checking. Moreover, given a user query, we show how the output from these different models can be combined to cross-examine the outputs of each other. Particularly, we show how accuracy in dialogue can be improved using the question answering model. We are also releasing all models used in the demo as a contribution of this paper. A short video demonstrating the system is available at https://ibm.box.com/v/emnlp2022-demo.

Nandana Mihindukulasooriya, Mike Sava, Gaetano Rossiello et al. · ibm-research

A research division plays an important role of driving innovation in an organization. Drawing insights, following trends, keeping abreast of new research, and formulating strategies are increasingly becoming more challenging for both researchers and executives as the amount of information grows in both velocity and volume. In this paper we present a use case of how a corporate research community, IBM Research, utilizes Semantic Web technologies to induce a unified Knowledge Graph from both structured and textual data obtained by integrating various applications used by the community related to research projects, academic papers, datasets, achievements and recognition. In order to make the Knowledge Graph more accessible to application developers, we identified a set of common patterns for exploiting the induced knowledge and exposed them as APIs. Those patterns were born out of user research which identified the most valuable use cases or user pain points to be alleviated. We outline two distinct scenarios: recommendation and analytics for business use. We will discuss these scenarios in detail and provide an empirical evaluation on entity recommendation specifically. The methodology used and the lessons learned from this work can be applied to other organizations facing similar challenges.

Michael Glass, Xueqing Wu, Ankita Rajaram Naik et al.

Data preparation, also called data wrangling, is considered one of the most expensive and time-consuming steps when performing analytics or building machine learning models. Preparing data typically involves collecting and merging data from complex heterogeneous, and often large-scale data sources, such as data lakes. In this paper, we introduce a novel approach toward automatic data wrangling in an attempt to alleviate the effort of end-users, e.g. data analysts, in structuring dynamic views from data lakes in the form of tabular data. We aim to address table augmentation tasks, including row/column population and data imputation. Given a corpus of tables, we propose a retrieval augmented self-trained transformer model. Our self-learning strategy consists in randomly ablating tables from the corpus and training the retrieval-based model to reconstruct the original values or headers given the partial tables as input. We adopt this strategy to first train the dense neural retrieval model encoding table-parts to vectors, and then the end-to-end model trained to perform table augmentation tasks. We test on EntiTables, the standard benchmark for table augmentation, as well as introduce a new benchmark to advance further research: WebTables. Our model consistently and substantially outperforms both supervised statistical methods and the current state-of-the-art transformer-based models.

Agostino Capponi, Alfio Gliozzo, Brian Zhu

Prediction markets allow users to trade on outcomes of real-world events, but are prone to fragmentation through overlapping questions, implicit equivalences, and hidden contradictions across markets. We present an agentic AI pipeline that autonomously (i) clusters markets into coherent topical groups using natural-language understanding over contract text and metadata, and (ii) identifies within-cluster market pairs whose resolved outcomes exhibit strong dependence, including same-outcome (correlated) and different-outcome (anti-correlated) relationships. Using a historical dataset of resolved markets on Polymarket, we evaluate the accuracy of the agent's relational predictions. We then translate discovered relationships into a simple trading strategy to quantify how these relationships map to actionable signals. Results show that agent-identified relationships achieve roughly 60-70% accuracy, and their induced trading strategies earn about 20% average returns over week-long horizons, highlighting the ability of agentic AI and large language models to uncover latent semantic structure in prediction markets.

Abhinav Goel, Agostino Capponi, Alfio Gliozzo et al.

We introduce SmartEval, a benchmark for systematically evaluating the quality of Solidity smart contracts generated by large language models (LLMs) from natural language specifications. SmartEval provides a corpus of 9,000 generated contracts paired with expert-written ground-truth implementations drawn from the FSMSCG dataset, a five-dimensional evaluation rubric covering functional completeness, variable fidelity, state-machine correctness, business-logic fidelity, and code quality, and a reproducible generation-and-evaluation pipeline. To validate the benchmark's reliability, we conduct three independent empirical studies: a five-condition ablation study (N=300 per condition) isolating the contribution of each pipeline component, a human expert evaluation by three Columbia University PhD researchers confirming automated scores align with expert judgment to within 0.34 points, and external security analysis via the Slither static analyzer confirming 79.4% agreement between the LLM auditor and a non-LLM rule-based tool. Systematic analysis of 9,000 generated contracts reveals characteristic failure modes (logic omissions at 35.3%, state transition errors at 23.4%, and complexity-driven degradation) and quantifies a +8.29 composite-score advantage of generated contracts over ground-truth implementations, attributable to LLMs' literal specification-following behavior. SmartEval establishes a reproducible, validated foundation for empirical research on LLM smart contract synthesis quality, with all data, evaluation code, and generated contracts publicly released.

Alfio Gliozzo, Naweed Khan, Christodoulos Constantinides et al.

This paper introduces Agentics, a functional agentic AI framework for building LLM-based structured data workflow pipelines. Designed for both research and practical applications, Agentics offers a new data-centric paradigm in which agents are embedded within data types, enabling logical transduction between structured states. This design shifts the focus toward principled data modeling, providing a declarative language where data types are directly exposed to large language models and composed through transductions triggered by type connections. We present a range of structured data workflow tasks and empirical evidence demonstrating the effectiveness of this approach, including data wrangling, text-to-SQL semantic parsing, and domain-specific multiple-choice question answering. The open source Agentics is available at https://github.com/IBM/Agentics.

Michael Glass, Gaetano Rossiello, Alfio Gliozzo

The ability to automatically extract Knowledge Graphs (KG) from a given collection of documents is a long-standing problem in Artificial Intelligence. One way to assess this capability is through the task of slot filling. Given an entity query in form of [Entity, Slot, ?], a system is asked to `fill' the slot by generating or extracting the missing value from a relevant passage or passages. This capability is crucial to create systems for automatic knowledge base population, which is becoming in ever-increasing demand, especially in enterprise applications. Recently, there has been a promising direction in evaluating language models in the same way we would evaluate knowledge bases, and the task of slot filling is the most suitable to this intent. The recent advancements in the field try to solve this task in an end-to-end fashion using retrieval-based language models. Models like Retrieval Augmented Generation (RAG) show surprisingly good performance without involving complex information extraction pipelines. However, the results achieved by these models on the two slot filling tasks in the KILT benchmark are still not at the level required by real-world information extraction systems. In this paper, we describe several strategies we adopted to improve the retriever and the generator of RAG in order to make it a better slot filler. Our KGI0 system (available at https://github.com/IBM/retrieve-write-slot-filling) reached the top-1 position on the KILT leaderboard on both T-REx and zsRE dataset with a large margin.

Michael Glass, Mustafa Eyceoz, Dharmashankar Subramanian et al.

Text-to-SQL is emerging as a practical interface for real world databases. The dominant paradigm for Text-to-SQL is cross-database or schema-independent, supporting application schemas unseen during training. The schema of a database defines the tables, columns, column types and foreign key connections between tables. Real world schemas can be large, containing hundreds of columns, but for any particular query only a small fraction will be relevant. Placing the entire schema in the prompt for an LLM can be impossible for models with smaller token windows and expensive even when the context window is large enough to allow it. Even apart from computational considerations, the accuracy of the model can be improved by focusing the SQL generation on only the relevant portion of the database. Schema linking identifies the portion of the database schema useful for the question. Previous work on schema linking has used graph neural networks, generative LLMs, and cross encoder classifiers. We introduce a new approach to adapt decoder-only LLMs to schema linking that is both computationally more efficient and more accurate than the generative approach. Additionally our extractive approach permits fine-grained control over the precision-recall trade-off for schema linking.

Jinheon Baek, Horst Samulowitz, Oktie Hassanzadeh et al. · ibm-research

Text-to-SQL aims to translate natural language queries into SQL statements, which is practical as it enables anyone to easily retrieve the desired information from databases. Recently, many existing approaches tackle this problem with Large Language Models (LLMs), leveraging their strong capability in understanding user queries and generating corresponding SQL code. Yet, the parametric knowledge in LLMs might be limited to covering all the diverse and domain-specific queries that require grounding in various database schemas, which makes generated SQLs less accurate oftentimes. To tackle this, we propose constructing the knowledge base for text-to-SQL, a foundational source of knowledge, from which we retrieve and generate the necessary knowledge for given queries. In particular, unlike existing approaches that either manually annotate knowledge or generate only a few pieces of knowledge for each query, our knowledge base is comprehensive, which is constructed based on a combination of all the available questions and their associated database schemas along with their relevant knowledge, and can be reused for unseen databases from different datasets and domains. We validate our approach on multiple text-to-SQL datasets, considering both the overlapping and non-overlapping database scenarios, where it outperforms relevant baselines substantially.

Agostino Capponi, Alfio Gliozzo, Chunghyun Han et al.

This paper presents a first empirical study of agentic AI as autonomous decision-makers in decentralized governance. Using more than 3K proposals from major protocols, we build an agentic AI voter that interprets proposal contexts, retrieves historical deliberation data, and independently determines its voting position. The agent operates within a realistic financial simulation environment grounded in verifiable blockchain data, implemented through a modular composable program (MCP) workflow that defines data flow and tool usage via Agentics framework. We evaluate how closely the agent's decisions align with the human and token-weighted outcomes, uncovering strong alignments measured by carefully designed evaluation metrics. Our findings demonstrate that agentic AI can augment collective decision-making by producing interpretable, auditable, and empirically grounded signals in realistic DAO governance settings. The study contributes to the design of explainable and economically rigorous AI agents for decentralized financial systems.

Feifei Pan, Mustafa Canim, Michael Glass et al.

Most existing end-to-end Table Question Answering (Table QA) models consist of a two-stage framework with a retriever to select relevant table candidates from a corpus and a reader to locate the correct answers from table candidates. Even though the accuracy of the reader models is significantly improved with the recent transformer-based approaches, the overall performance of such frameworks still suffers from the poor accuracy of using traditional information retrieval techniques as retrievers. To alleviate this problem, we introduce T-RAG, an end-to-end Table QA model, where a non-parametric dense vector index is fine-tuned jointly with BART, a parametric sequence-to-sequence model to generate answer tokens. Given any natural language question, T-RAG utilizes a unified pipeline to automatically search through a table corpus to directly locate the correct answer from the table cells. We apply T-RAG to recent open-domain Table QA benchmarks and demonstrate that the fine-tuned T-RAG model is able to achieve state-of-the-art performance in both the end-to-end Table QA and the table retrieval tasks.

Jian Ni, Gaetano Rossiello, Alfio Gliozzo et al.

Relation extraction (RE) is an important information extraction task which provides essential information to many NLP applications such as knowledge base population and question answering. In this paper, we present a novel generative model for relation extraction and classification (which we call GREC), where RE is modeled as a sequence-to-sequence generation task. We explore various encoding representations for the source and target sequences, and design effective schemes that enable GREC to achieve state-of-the-art performance on three benchmark RE datasets. In addition, we introduce negative sampling and decoding scaling techniques which provide a flexible tool to tune the precision and recall performance of the model. Our approach can be extended to extract all relation triples from a sentence in one pass. Although the one-pass approach incurs certain performance loss, it is much more computationally efficient.

Md Faisal Mahbub Chowdhury, Gaetano Rossiello, Michael Glass et al.

In recent years, a number of keyphrase generation (KPG) approaches were proposed consisting of complex model architectures, dedicated training paradigms and decoding strategies. In this work, we opt for simplicity and show how a commonly used seq2seq language model, BART, can be easily adapted to generate keyphrases from the text in a single batch computation using a simple training procedure. Empirical results on five benchmarks show that our approach is as good as the existing state-of-the-art KPG systems, but using a much simpler and easy to deploy framework.

Nandana Mihindukulasooriya, Mohnish Dubey, Alfio Gliozzo et al.

Each year the International Semantic Web Conference organizes a set of Semantic Web Challenges to establish competitions that will advance state-of-the-art solutions in some problem domains. The Semantic Answer Type and Relation Prediction Task (SMART) task is one of the ISWC 2021 Semantic Web challenges. This is the second year of the challenge after a successful SMART 2020 at ISWC 2020. This year's version focuses on two sub-tasks that are very important to Knowledge Base Question Answering (KBQA): Answer Type Prediction and Relation Prediction. Question type and answer type prediction can play a key role in knowledge base question answering systems providing insights about the expected answer that are helpful to generate correct queries or rank the answer candidates. More concretely, given a question in natural language, the first task is, to predict the answer type using a target ontology (e.g., DBpedia or Wikidata. Similarly, the second task is to identify relations in the natural language query and link them to the relations in a target ontology. This paper discusses the task descriptions, benchmark datasets, and evaluation metrics. For more information, please visit https://smart-task.github.io/2021/.

Saneem Ahmed Chemmengath, Vishwajeet Kumar, Samarth Bharadwaj et al.

Weakly-supervised table question-answering(TableQA) models have achieved state-of-art performance by using pre-trained BERT transformer to jointly encoding a question and a table to produce structured query for the question. However, in practical settings TableQA systems are deployed over table corpora having topic and word distributions quite distinct from BERT's pretraining corpus. In this work we simulate the practical topic shift scenario by designing novel challenge benchmarks WikiSQL-TS and WikiTQ-TS, consisting of train-dev-test splits in five distinct topic groups, based on the popular WikiSQL and WikiTableQuestions datasets. We empirically show that, despite pre-training on large open-domain text, performance of models degrades significantly when they are evaluated on unseen topics. In response, we propose T3QA (Topic Transferable Table Question Answering) a pragmatic adaptation framework for TableQA comprising of: (1) topic-specific vocabulary injection into BERT, (2) a novel text-to-text transformer generator (such as T5, GPT2) based natural language question generation pipeline focused on generating topic specific training data, and (3) a logical form reranker. We show that T3QA provides a reasonably good baseline for our topic shift benchmarks. We believe our topic split benchmarks will lead to robust TableQA solutions that are better suited for practical deployment.

Michael Glass, Gaetano Rossiello, Md Faisal Mahbub Chowdhury et al.

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.

Gaetano Rossiello, Nandana Mihindukulasooriya, Ibrahim Abdelaziz et al.

Relation linking is essential to enable question answering over knowledge bases. Although there are various efforts to improve relation linking performance, the current state-of-the-art methods do not achieve optimal results, therefore, negatively impacting the overall end-to-end question answering performance. In this work, we propose a novel approach for relation linking framing it as a generative problem facilitating the use of pre-trained sequence-to-sequence models. We extend such sequence-to-sequence models with the idea of infusing structured data from the target knowledge base, primarily to enable these models to handle the nuances of the knowledge base. Moreover, we train the model with the aim to generate a structured output consisting of a list of argument-relation pairs, enabling a knowledge validation step. We compared our method against the existing relation linking systems on four different datasets derived from DBpedia and Wikidata. Our method reports large improvements over the state-of-the-art while using a much simpler model that can be easily adapted to different knowledge bases.

Yannis Katsis, Saneem Chemmengath, Vishwajeet Kumar et al.

Recent advances in transformers have enabled Table Question Answering (Table QA) systems to achieve high accuracy and SOTA results on open domain datasets like WikiTableQuestions and WikiSQL. Such transformers are frequently pre-trained on open-domain content such as Wikipedia, where they effectively encode questions and corresponding tables from Wikipedia as seen in Table QA dataset. However, web tables in Wikipedia are notably flat in their layout, with the first row as the sole column header. The layout lends to a relational view of tables where each row is a tuple. Whereas, tables in domain-specific business or scientific documents often have a much more complex layout, including hierarchical row and column headers, in addition to having specialized vocabulary terms from that domain. To address this problem, we introduce the domain-specific Table QA dataset AIT-QA (Airline Industry Table QA). The dataset consists of 515 questions authored by human annotators on 116 tables extracted from public U.S. SEC filings (publicly available at: https://www.sec.gov/edgar.shtml) of major airline companies for the fiscal years 2017-2019. We also provide annotations pertaining to the nature of questions, marking those that require hierarchical headers, domain-specific terminology, and paraphrased forms. Our zero-shot baseline evaluation of three transformer-based SOTA Table QA methods - TaPAS (end-to-end), TaBERT (semantic parsing-based), and RCI (row-column encoding-based) - clearly exposes the limitation of these methods in this practical setting, with the best accuracy at just 51.8\% (RCI). We also present pragmatic table preprocessing steps used to pivot and project these complex tables into a layout suitable for the SOTA Table QA models.

Feifei Pan, Mustafa Canim, Michael Glass et al.

We present the first end-to-end, transformer-based table question answering (QA) system that takes natural language questions and massive table corpus as inputs to retrieve the most relevant tables and locate the correct table cells to answer the question. Our system, CLTR, extends the current state-of-the-art QA over tables model to build an end-to-end table QA architecture. This system has successfully tackled many real-world table QA problems with a simple, unified pipeline. Our proposed system can also generate a heatmap of candidate columns and rows over complex tables and allow users to quickly identify the correct cells to answer questions. In addition, we introduce two new open-domain benchmarks, E2E_WTQ and E2E_GNQ, consisting of 2,005 natural language questions over 76,242 tables. The benchmarks are designed to validate CLTR as well as accommodate future table retrieval and end-to-end table QA research and experiments. Our experiments demonstrate that our system is the current state-of-the-art model on the table retrieval task and produces promising results for end-to-end table QA.

Michael Glass, Mustafa Canim, Alfio Gliozzo et al.

Transformer based architectures are recently used for the task of answering questions over tables. In order to improve the accuracy on this task, specialized pre-training techniques have been developed and applied on millions of open-domain web tables. In this paper, we propose two novel approaches demonstrating that one can achieve superior performance on table QA task without even using any of these specialized pre-training techniques. The first model, called RCI interaction, leverages a transformer based architecture that independently classifies rows and columns to identify relevant cells. While this model yields extremely high accuracy at finding cell values on recent benchmarks, a second model we propose, called RCI representation, provides a significant efficiency advantage for online QA systems over tables by materializing embeddings for existing tables. Experiments on recent benchmarks prove that the proposed methods can effectively locate cell values on tables (up to ~98% Hit@1 accuracy on WikiSQL lookup questions). Also, the interaction model outperforms the state-of-the-art transformer based approaches, pre-trained on very large table corpora (TAPAS and TaBERT), achieving ~3.4% and ~18.86% additional precision improvement on the standard WikiSQL benchmark.

Sarthak Dash, Nandana Mihindukulasooriya, Alfio Gliozzo et al.

Inferring semantic types for entity mentions within text documents is an important asset for many downstream NLP tasks, such as Semantic Role Labelling, Entity Disambiguation, Knowledge Base Question Answering, etc. Prior works have mostly focused on supervised solutions that generally operate on relatively small-to-medium-sized type systems. In this work, we describe two systems aimed at predicting type information for the following two tasks, namely, a TypeSuggest module, an unsupervised system designed to predict types for a set of user-entered query terms, and an Answer Type prediction module, that provides a solution for the task of determining the correct type of the answer expected to a given query. Our systems generalize to arbitrary type systems of any sizes, thereby making it a highly appealing solution to extract type information at any granularity.

Sarthak Dash, Gaetano Rossiello, Nandana Mihindukulasooriya et al.

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), 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.

Pavan Kapanipathi, Ibrahim Abdelaziz, Srinivas Ravishankar et al.

Knowledge base question answering (KBQA)is an important task in Natural Language Processing. Existing approaches face significant challenges including complex question understanding, necessity for reasoning, and lack of large end-to-end training datasets. In this work, we propose Neuro-Symbolic Question Answering (NSQA), a modular KBQA system, that leverages (1) Abstract Meaning Representation (AMR) parses for task-independent question understanding; (2) a simple yet effective graph transformation approach to convert AMR parses into candidate logical queries that are aligned to the KB; (3) a pipeline-based approach which integrates multiple, reusable modules that are trained specifically for their individual tasks (semantic parser, entity andrelationship linkers, and neuro-symbolic reasoner) and do not require end-to-end training data. NSQA achieves state-of-the-art performance on two prominent KBQA datasets based on DBpedia (QALD-9 and LC-QuAD1.0). Furthermore, our analysis emphasizes that AMR is a powerful tool for KBQA systems.

Nandana Mihindukulasooriya, Mohnish Dubey, Alfio Gliozzo et al.

Each year the International Semantic Web Conference accepts a set of Semantic Web Challenges to establish competitions that will advance the state of the art solutions in any given problem domain. The SeMantic AnsweR Type prediction task (SMART) was part of ISWC 2020 challenges. Question type and answer type prediction can play a key role in knowledge base question answering systems providing insights that are helpful to generate correct queries or rank the answer candidates. More concretely, given a question in natural language, the task of SMART challenge is, to predict the answer type using a target ontology (e.g., DBpedia or Wikidata).

Nandana Mihindukulasooriya, Gaetano Rossiello, Pavan Kapanipathi et al.

Knowledgebase question answering systems are heavily dependent on relation extraction and linking modules. However, the task of extracting and linking relations from text to knowledgebases faces two primary challenges; the ambiguity of natural language and lack of training data. To overcome these challenges, we present SLING, a relation linking framework which leverages semantic parsing using Abstract Meaning Representation (AMR) and distant supervision. SLING integrates multiple relation linking approaches that capture complementary signals such as linguistic cues, rich semantic representation, and information from the knowledgebase. The experiments on relation linking using three KBQA datasets; QALD-7, QALD-9, and LC-QuAD 1.0 demonstrate that the proposed approach achieves state-of-the-art performance on all benchmarks.

Sarthak Dash, Md Faisal Mahbub Chowdhury, Alfio Gliozzo et al.

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.

Lin Pan, Rishav Chakravarti, Anthony Ferritto et al.

Existing literature on Question Answering (QA) mostly focuses on algorithmic novelty, data augmentation, or increasingly large pre-trained language models like XLNet and RoBERTa. Additionally, a lot of systems on the QA leaderboards do not have associated research documentation in order to successfully replicate their experiments. In this paper, we outline these algorithmic components such as Attention-over-Attention, coupled with data augmentation and ensembling strategies that have shown to yield state-of-the-art results on benchmark datasets like SQuAD, even achieving super-human performance. Contrary to these prior results, when we evaluate on the recently proposed Natural Questions benchmark dataset, we find that an incredibly simple approach of transfer learning from BERT outperforms the previous state-of-the-art system trained on 4 million more examples than ours by 1.9 F1 points. Adding ensembling strategies further improves that number by 2.3 F1 points.

Michael Glass, Alfio Gliozzo, Rishav Chakravarti et al.

BERT (Bidirectional Encoder Representations from Transformers) and related pre-trained Transformers have provided large gains across many language understanding tasks, achieving a new state-of-the-art (SOTA). BERT is pre-trained on two auxiliary tasks: Masked Language Model and Next Sentence Prediction. In this paper we introduce a new pre-training task inspired by reading comprehension to better align the pre-training from memorization to understanding. Span Selection Pre-Training (SSPT) poses cloze-like training instances, but rather than draw the answer from the model's parameters, it is selected from a relevant passage. We find significant and consistent improvements over both BERT-BASE and BERT-LARGE on multiple reading comprehension (MRC) datasets. Specifically, our proposed model has strong empirical evidence as it obtains SOTA results on Natural Questions, a new benchmark MRC dataset, outperforming BERT-LARGE by 3 F1 points on short answer prediction. We also show significant impact in HotpotQA, improving answer prediction F1 by 4 points and supporting fact prediction F1 by 1 point and outperforming the previous best system. Moreover, we show that our pre-training approach is particularly effective when training data is limited, improving the learning curve by a large amount.

Sarthak Dash, Michael R. Glass, Alfio Gliozzo et al.

In this paper, we propose a fully automated system to extend knowledge graphs using external information from web-scale corpora. The designed system leverages a deep learning based technology for relation extraction that can be trained by a distantly supervised approach. In addition to that, the system uses a deep learning approach for knowledge base completion by utilizing the global structure information of the induced KG to further refine the confidence of the newly discovered relations. The designed system does not require any effort for adaptation to new languages and domains as it does not use any hand-labeled data, NLP analytics and inference rules. Our experiments, performed on a popular academic benchmark demonstrate that the suggested system boosts the performance of relation extraction by a wide margin, reporting error reductions of 50%, resulting in relative improvement of up to 100%. Also, a web-scale experiment conducted to extend DBPedia with knowledge from Common Crawl shows that our system is not only scalable but also does not require any adaptation cost, while yielding substantial accuracy gain.

Sarthak Dash, Alfio Gliozzo

State-of-the-art approaches for Knowledge Base Completion (KBC) exploit deep neural networks trained with both false and true assertions: positive assertions are explicitly taken from the knowledge base, whereas negative ones are generated by random sampling of entities. In this paper, we argue that random sampling is not a good training strategy since it is highly likely to generate a huge number of nonsensical assertions during training, which does not provide relevant training signal to the system. Hence, it slows down the learning process and decreases accuracy. To address this issue, we propose an alternative approach called Distributional Negative Sampling that generates meaningful negative examples which are highly likely to be false. Our approach achieves a significant improvement in Mean Reciprocal Rank values amongst two different KBC algorithms in three standard academic benchmarks.