LAMRF: Logic-adaptive multi-source reasoning fusion for multi-hop knowledge graph question answering

Multi-hop question answering (QA) requires a model to retrieve and integrate information from multiple passages to answer a question. Rapid progress has been made on multi-hop QA systems with regard to standard evaluation metrics, including EM and F1. However, by simply evaluating the correctness of the answers, it is unclear to what extent these systems have learned the ability to perform multi-hop reasoning. In this paper, we propose an additional sub-question evaluation for the multi-hop QA dataset HotpotQA, in order to shed some light on explaining the reasoning process of QA systems in answering complex questions. We adopt a neural decomposition model to generate sub-questions for a multi-hop question, followed by extracting the corresponding sub-answers. Contrary to our expectation, multiple state-of-the-art multi-hop QA models fail to answer a large portion of sub-questions, although the corresponding multi-hop questions are correctly answered. Our work takes a step forward towards building a more explainable multi-hop QA system.

Given unstructured text, Large Language Models (LLMs) are adept at answering simple (single-hop) questions. However, as the complexity of the questions increase, the performance of LLMs degrade. We believe this is due to the overhead associated with understanding the complex question followed by filtering and aggregating unstructured information in the raw text. Recent methods try to reduce this burden by integrating structured knowledge triples into the raw text, aiming to provide a structured overview that simplifies information processing. However, this simplistic approach is query-agnostic and the extracted facts are ambiguous as they lack context. To address these drawbacks and to enable LLMs to answer complex (multi-hop) questions with ease, we propose to use a knowledge graph (KG) that is context-aware and is distilled to contain query-relevant information. The use of our compressed distilled KG as input to the LLM results in our method utilizing up to 67% fewer tokens to represent the query relevant information present in the supporting documents, compared to the state-of-the-art (SoTA) method. Our experiments show consistent improvements over the SoTA across several metrics (EM, F1, BERTScore, and Human Eval) on two popular benchmark datasets (HotpotQA and MuSiQue).

Question generation (QG) attempts to solve the inverse of question answering (QA) problem by generating a natural language question given a document and an answer. While sequence to sequence neural models surpass rule-based systems for QG, they are limited in their capacity to focus on more than one supporting fact. For QG, we often require multiple supporting facts to generate high-quality questions. Inspired by recent works on multi-hop reasoning in QA, we take up Multi-hop question generation, which aims at generating relevant questions based on supporting facts in the context. We employ multitask learning with the auxiliary task of answer aware supporting fact prediction to guide the question generator. In addition, we also proposed a question-aware reward function in a Reinforcement Learning (RL) framework to maximize the utilization of the supporting facts. We demonstrate the effectiveness of our approach through experiments on the multi-hop question answering dataset, HotPotQA. Empirical evaluation shows our model to outperform the single-hop neural question generation models on both automatic evaluation metrics such as BLEU, METEOR, and ROUGE and human evaluation metrics for quality and coverage of the generated questions.

Multi-hop questions are associated with a series of justifications, and one needs to obtain the answers by following the reasoning path (RP) that orders the justifications adequately. So reasoning path retrieval becomes a critical preliminary stage for multi-hop Question Answering (QA). Within the RP, two fundamental challenges emerge for better performance: (i) what the order of the justifications in the RP should be, and (ii) what if the wrong justification has been in the path. In this paper, we propose Reasoning Path Augmentation (RPA), which uses reasoning path reordering and augmentation to handle the above two challenges, respectively. Reasoning path reordering restructures the reasoning by targeting the easier justification first but difficult one later, in which the difficulty is determined by the overlap between query and justifications since the higher overlap means more lexical relevance and easier searchable. Reasoning path augmentation automatically generates artificial RPs, in which the distracted justifications are inserted to aid the model recover from the wrong justification. We build RPA with a naive pre-trained model and evaluate RPA on the QASC and MultiRC datasets. The evaluation results demonstrate that RPA outperforms previously published reasoning path retrieval methods, showing the effectiveness of the proposed methods. Moreover, we present detailed experiments on how the orders of justifications and the percent of augmented paths affect the question- answering performance, revealing the importance of polishing RPs and the necessity of augmentation.

Knowledge editing aims to update Large Language Models (LLMs) with new information without costly retraining.However, consistently reflecting these updates in complex multihop Question Answering (QA), which demands reasoning over interconnected facts, is challenging.Many existing methods overlook the interplay with pre-existing knowledge, leading to inconsistent edit propagation.To overcome this, we introduce StepKE (Stepwise Knowledge Editing for Multi-hop QA), a novel framework for robustly integrating edited and existing knowledge for coherent multi-hop reasoning.StepKE uniquely decomposes multihop questions into sequential single-hop subquestions, retrieving relevant facts (both edited and pre-existing) from an external knowledge graph for each step.It employs context-aware prompting with prior reasoning history and finetuning for precise edit propagation.This systematic integration enables effective stepwise reasoning.Experiments show StepKE generates significantly more accurate and consistent responses than baselines, showcasing strong knowledge editing and integration in multi-hop QA.

Multi-hop question generation (MQG) aims to generate complex questions which require reasoning over multiple pieces of information of the input passage. Most existing work on MQG has focused on exploring graph-based networks to equip the traditional Sequence-to-sequence framework with reasoning ability. However, these models do not take full advantage of the constraint between questions and answers. Furthermore, studies on multi-hop question answering (QA) suggest that Transformers can replace the graph structure for multi-hop reasoning. Therefore, in this work, we propose a novel framework, QA4QG, a QA-augmented BART-based framework for MQG. It augments the standard BART model with an additional multi-hop QA module to further constrain the generated question. Our results on the HotpotQA dataset show that QA4QG outperforms all state-of-the-art models, with an increase of 8 BLEU-4 and 8 ROUGE points compared to the best results previously reported. Our work suggests the advantage of introducing pre-trained language models and QA module for the MQG task.

Multimodal Knowledge Editing (MKE) extends traditional knowledge editing to settings involving both textual and visual modalities. However, existing MKE benchmarks primarily assess final answer correctness, neglecting the quality of intermediate reasoning and robustness to visually rephrased inputs. To address this limitation, we introduce MMQAKE, the first benchmark for multimodal multihop question answering with knowledge editing. MMQAKE evaluates: (1) a model’s ability to reason over 2–5-hop factual chains that span both text and images, including performance at each intermediate step; (2) robustness to visually rephrased inputs in multihop questions. Our evaluation shows that current MKE methods often struggle to consistently update and reason over multimodal reasoning chains following knowledge edits. To overcome these challenges, we propose Hybrid-DMKG, a hybrid reasoning framework built on a dynamic multimodal knowledge graph (DMKG) to enable accurate multihop reasoning over updated multimodal knowledge. Hybrid-DMKG first uses a large language model to decompose multimodal multihop questions into sequential sub-questions, then applies a multimodal retrieval model to locate updated facts by jointly encoding each sub-question with candidate entities and their associated images. For answer inference, a hybrid reasoning module operates over the DMKG via two parallel paths: (1) relation-linking prediction; (2) RAG Reasoning with large vision-language models. A background-reflective decision module then aggregates evidence from both paths to select the most credible answer. Experimental results on MMQAKE show that Hybrid-DMKG significantly outperforms existing MKE approaches, achieving higher accuracy and improved robustness to knowledge updates.

Answering multi-hop questions requires first retrieving a sequence of supporting facts, and the order in which these facts are retrieved significantly affects retriever performance. To achieve a clearer reasoning order, it is beneficial to address the easier facts first then move to the more difficult ones. However, current orders are usually pre-defined during data construction or specified manually, which restricts the model’s reasoning potential. This paper proposes Auto-Optimizing Reasoning Order (AORO), a method to automatically optimize the reasoning order for each sample, where difficulty is determined by a retrieval model trained with carefully curated data. First, a retriever is trained using data that encompasses all combinations of the possible reasoning orders. The trained retriever is then used to assess the difficulty of each fact, placing the fact with the least difficulty at the beginning of the sequence. Next, the retrieval model is retrained based on these optimized sequences, which are empirically better suited to its capabilities. This process creates an iterative self-debiasing paradigm, and these steps are repeated until all facts are reordered. Experiments conducted on two multi-hop QA benchmarks, QASC and MultiRC, demonstrate the effectiveness of AORO, which outperforms strong baselines using the same PTM, and further enables advanced PTMs to achieve improvements of up to 1.6 points in Recall@10 and 3.7 points in F1 score. Additional case analyses reveal empirical patterns in the optimal reasoning order: the pattern appears independent of the dataset and the underlying pre-trained model; and the sequence proceeds by confirming the truth of the question, answering the question, and filling in any gaps, which aligns with human reasoning.

Large Language Models (LLMs) have shown proficiency in question-answering tasks but often struggle to integrate real-time knowledge, leading to potentially outdated or inaccurate responses. This problem becomes even more challenging when dealing with multi-hop questions, since they require LLMs to update and integrate multiple knowledge pieces relevant to the questions. To tackle the problem, we propose the Retrieval-Augmented model Editing (RAE) framework for multi-hop question answering. RAE first retrieves edited facts and then refines the language model through in-context learning. Specifically, our retrieval approach, based on mutual information maximization, leverages the reasoning abilities of LLMs to identify chain facts that traditional similarity-based searches might miss. In addition, our framework includes a pruning strategy to eliminate redundant information from the retrieved facts, which enhances the editing accuracy and mitigates the hallucination problem. Our framework is supported by theoretical justification for its fact retrieval efficacy. Finally, comprehensive evaluation across various LLMs validates RAE's ability in providing accurate answers with updated knowledge. Our code is available at: https://github.com/sycny/RAE.

The target of the multi-hop knowledge base question-answering task is to find answers of some factoid questions by reasoning across multiple knowledge triples in the knowledge base. Most of the existing methods for multi-hop knowledge base question answering based on a general knowledge graph ignore the semantic relationship between each hop. However, modeling the knowledge base as a directed hypergraph has the problems of sparse incidence matrices and asymmetric Laplacian matrices. To make up for the deficiency, we propose a directed hypergraph convolutional network modeled on hyperbolic space, which can better deal with the sparse structure, and effectively adapt to the problem of an asymmetric incidence matrix of directed hypergraphs modeled on a knowledge base. We propose an interpretable KBQA model based on the hyperbolic directed hypergraph convolutional neural network named HDH-GCN which can update relation semantic information hop-by-hop and pays attention to different relations at different hops. The model can improve the accuracy of the multi-hop knowledge base question-answering task, and has application value in text question answering, human–computer interactions and other fields. Extensive experiments on benchmarks—PQL, MetaQA—demonstrate the effectiveness and universality of our HDH-GCN model, leading to state-of-the-art performance.

In today’s fast-paced industry, professionals face the challenge of summarizing a large number of documents and extracting vital information from them on a daily basis. These metrics are frequently hidden away in tables and/or their nested hyperlinks. To address this challenge, the approach of Table Question Answering (QA) has been developed to extract the relevant information. However, traditional Table QA training tasks that provide a table and an answer(s) from a gold cell coordinate(s) for a question may not always ensure extracting the accurate answer(s). Recent advancements in Large Language Models (LLMs) have opened up new possibilities for extracting information from tabular data using prompts. In this paper, we introduce the Multi-hop Few-shot Open Rich Table QA (MFORT-QA) approach, which consists of two major steps. The first step involves Few-Shot Learning (FSL), where relevant tables and associated contexts of hyperlinks are retrieved based on a given question. The retrieved content is then used to construct few-shot prompts as inputs to an LLM, such as ChatGPT. To tackle the challenge of answering complex questions, the second step leverages Chain-of-thought (CoT) prompting to decompose the complex question into a sequential chain of questions and reasoning thoughts in a multi-hop manner. Retrieval-Augmented Generation (RAG) enhances this process by retrieving relevant tables and contexts of hyperlinks that are relevant to the resulting reasoning thoughts and questions. These additional contexts are then used to supplement the prompt used in the first step, resulting in more accurate answers from an LLM. Empirical results from OTT-QA demonstrate that our abstractive QA approach significantly improves the accuracy of extractive Table QA methods.

Explainability and interpretability are two important concepts, the absence of which can and should impede the application of well-performing neural networks to real-world problems. At the same time, they are difficult to incorporate into the large, black-box models that achieve state-of-the-art results in a multitude of NLP tasks. Bidirectional Encoder Representations from Transformers (BERT) is one such black-box model. It has become a staple architecture to solve many different NLP tasks and has inspired a number of related Transformer models. Understanding how these models draw conclusions is crucial for both their improvement and application. We contribute to this challenge by presenting VisBERT, a tool for visualizing the contextual token representations within BERT for the task of (multi-hop) Question Answering. Instead of analyzing attention weights, we focus on the hidden states resulting from each encoder block within the BERT model. This way we can observe how the semantic representations are transformed throughout the layers of the model. VisBERT enables users to get insights about the model’s internal state and to explore its inference steps or potential shortcomings. The tool allows us to identify distinct phases in BERT’s transformations that are similar to a traditional NLP pipeline and offer insights during failed predictions.

The advance on Question Answering systems has achieved significant results, and new related problems have arisen. Few studies have addressed the Complex Knowledge Base Question Answering (C-KBQA) problem. This work presents an overview of C-KBQA systems. A collection of 54 papers were systematically selected, and a map of the methods, approaches, trends, future directions, and gaps in the C-KBQA research was performed. This study shows that C-KBQA systems need to handle multi-hop and constraint questions, and two approaches are usually used to address this problem.

Retrieval-augmented generation (RAG) offers an effective approach for addressing question answering (QA) tasks. However, the imperfections of the retrievers in RAG models often result in the retrieval of irrelevant information, which could introduce noise and degrade the performance, especially when handling multi-hop questions that require multiple steps of reasoning. To enhance the multi-hop reasoning ability of RAG models, we propose TRACE1. TRACE constructs knowledge-grounded reasoning chains, which are a series of logically connected knowledge triples, to identify and integrate supporting evidence from the retrieved documents for answering questions. Specifically, TRACE employs a KG Generator to create a knowledge graph (KG) from the retrieved documents, and then uses a novel Autoregressive Reasoning Chain Constructor to build reasoning chains. Experimental results on three multi-hop QA datasets show that TRACE achieves an average performance improvement of up to 14.03% compared to using all the retrieved documents. Moreover, the results indicate that using reasoning chains as context, rather than the entire documents, is often sufficient to correctly answer questions.

Over the past decade, Knowledge Graphs (KG) have emerged as a prominent repository for storing facts about the world in a linked data architecture. Providing machines with the capability of exploring such Knowledge Graphs and answering natural language questions over them, has been an active area of research. The purpose of this work, is to delve further into the research of retrieving information stored in KGs, based on the natural language questions posed by the user. Knowledge Graph Question Answering (KGQA) aims to produce a concise answer to a user question, such that the user is exempt from using KG vocabulary and overheads of learning a formal query language. Existing KGQA systems have achieved excellent results over Simple Questions, where the information required is limited to a single triple and a single formal query pattern. Our motivation is to improve the performance of KGQA over Complex Questions, where formal query patterns significantly vary, and a single triple is not confining for all the required information. Complex KGQA provides several challenges such as understanding semantics and syntactic structure of questions, Entity Linking, Relation Linking and Answer Representation. Lack of suitable datasets for complex question answering further adds to research gaps. Hence, in this thesis, we focus the research objective of laying the foundations for the advancement of the state-of-the-art for Complex Question Answering over Knowledge Graphs, by providing techniques to solve various challenges and provide resources to fill the research gaps. First, we propose Normalized Query Structure (NQS), which is a linguistic analyzer module that helps the QA system to detect inputs and intents and the relation between them in the users' question. NQS acts like an intermediate language between natural language questions and formal expressions to ease the process of query formulation for complex questions. We then developed a framework named LC-QuAD to generate large scale question answering dataset by reversing the process of question answering, thereby translating natural language questions from the formal query using intermediate templates. Our goal is to use this framework for high variations in the query patterns and create a large size dataset with minimum human effort. The first version of the dataset consists of 5,000 complex questions. By extending the LC-QuAD framework to support Reified KGs and crowd-sourcing, we published the second version of the dataset as LC-QuAD 2.0, consisting of 30,000 questions with their paraphrases and has higher complexity and new variations in the questions. To overcome the problem of Entity Linking and Relation Linking in KGQA, we develop EARL, a module performing these two tasks as a single joint task for complex question answering. We develop approaches for this module, first by formalizing the task as an instance of the Generalized Traveling Salesman Problem (GTSP) and the second approach uses machine learning to exploit the connection density between nodes in the Knowledge Graph. Lastly, we create another large scale dataset to answer verbalization and provide results for multiple baseline system s on it. The Verbalization dataset is introduced to make the system's response more human-like. The NQS based KGQA system was next to the best system in terms of accuracy on the QALD-5 dataset. We empirically prove that NQS is robust to tackle paraphrases of the questions. EARL achieves the state of the art results in Entity Linking and Relation Linking for question answering on several KGQA datasets. The dataset curated in this thesis has helped the research community to move forward in the direction of improving the accuracy of complex question answering as a task as other researchers too developed several KGQA systems and modules around these published datasets. With the large-scale datasets, we have encouraged the use of large scale machine learning, deep learning and emergence of new techniques to advance the state-of-the-art in complex question answering over knowledge graphs. We further developed core components for the KGQA pipeline to overcome the challenges of Question Understanding, Entity-Relation Linking and Answer Verbalization and thus achieve our research objective. All the proposed approaches mentioned in this thesis and the published resources are available at https://github.com/AskNowQA and are released under the umbrella project AskNow.