Research on Multi-hop Q/A

Accurately generating graph queries based on natural language queries that can be executed against a knowledge graph. The technique involves using a large language model (LLM) to generate graph queries from natural language queries, validating the generated queries, and executing the validated queries against the knowledge graph. Errors in the generated queries are detected during validation and the LLM is provided feedback to generate modified queries that can be executed correctly. This improves the accuracy of results returned from the knowledge graph compared to running unmodified queries with errors.

Source

Improving the accuracy and efficiency of natural language processing (NLP) systems in generating answers to questions using knowledge-enhanced chain-of-thought prompting. The method involves generating natural language logic paths between the question and candidate answers based on a knowledge graph. These logic paths are then incorporated into the prompt provided to the NLP model to guide its reasoning and output selection. This allows leveraging structured knowledge from graphs to enhance chain-of-thought prompting for better accuracy and explainability compared to just using pretrained models.

Source

Modular machine learning pipeline for query resolution that breaks complex queries into independent sub-problems that can be answered and aggregated to process the query. The pipeline consists of three connected stages: (1) retrieval to extract relevant evidence passages, (2) aggregation to score and combine the passages, and (3) resolution using a language model to generate the query response. The retrieval and aggregation stages are trained partially independently and partially end-to-end to specialize for specific tasks while the resolution stage is trained jointly to generalize to new query types. This allows the pipeline to be tailored to the sub-problems while still being adaptable to new queries.

Source

Integrated system for managing, querying, and generating responses related to interconnected data using a graph database, specialized loaders, natural language reasoning, and a multi-modal user interface. The system ingests structured, unstructured, and time-series data from diverse sources, transforms it into a unified graph representation, and enables sophisticated reasoning tasks across modalities. It provides a secure, scalable, and interoperable solution for integrating, analyzing, visualizing, and reasoning over heterogeneous data at scale.

Source

Graph-based natural language processing system for querying, analyzing, and visualizing complex data structures using AI language models. The system trains an AI language model on input data from original sources using a training algorithm with extraction and condensing steps. It extracts data from sources, loads it into a graph database, transforms into a condensed natural language format, trains on graph structure, and responds to natural language questions.

Source

This paper presents a novel Retrieval-AugmentedGeneration (RAG) framework tailored for complex questionanswering tasks, addressing challenges in multi-hop reasoningand contextual understanding across lengthy documents. Builtupon LLaMA 3, the integrates dense retrievalmodule with advanced context fusion and reasoningmechanisms, enabling more accurate coherent responsegeneration. A joint optimization strategy combining retrievallikelihood generation cross-entropy improves modelsrobustness adaptability. Experimental results show that theproposed system outperforms existing retrieval-augmented andgenerative baselines, confirming its effectiveness deliveringprecise, contextually grounded answers.

Source

Generative, graph-enhanced information retrieval approaches for cluttered datasets that enable more accurate and explainable search results for complex questions. The approach uses graph representation of documents with nodes representing entities and edges connecting relationships. Embedding-based retrieval finds relevant nodes, then graph-based retrieval extracts a path connecting them. This path is fed to a generative AI model to generate a customized response. It improves over embedding-only retrieval for complex queries by connecting dispersed information and explaining how it was assembled.

Source

A method for improving question answering (QA) accuracy across different structured databases like knowledge graphs and tables by training a unified QA model. The model is trained on samples with questions, database type, and query instructions. It learns to generate query instructions from questions and database types. This allows the model to understand and query multiple structured databases instead of degenerating to unstructured text QA.

Source

Multi-Hop Knowledge Graph Question Answering (multi-hop KGQA) aims to obtain answers by analyzing the semantics of natural language questions and performing multi-step reasoning across multiple entities relations in knowledge graphs. Traditional embedding-based methods map graphs into vector spaces for answer matching through operations. While these approaches have improved model performance, they face two critical challenges: lack clear interpretability caused implicit mechanisms, semantic gap between queries structured representations. This study proposes DRKG (Decomposed Reasoning over Graph), a constrained multi-hop framework based on large models (LLMs) that introduces explicit plans as logical boundary controllers. The innovation lies key aspects: First, generates hop-constrained parsing LLMs, explicitly defining traversal path length entity-retrieval logic Second, conducts selective retrieval during graph plans, ensuring faithfulness knowledge. We evaluate four datasets, experimental results demonstrate achieves 1%5% accuracy improvements best baseline models. Additional ab... Read More

Source

Using large language models to answer table question-answering tasks while preserving data security and privacy by having the model generate code to access the table instead of directly seeing the table data. The model, called the Solver, receives a user query and generates code based on instructions and schema provided by an Oracle. The Oracle evaluates the code and may ask follow-up questions. This allows the model to answer questions without exposing the underlying table data. If the answer is incorrect, the Oracle provides guidance to help the model improve. The game-like interaction between the Solver and Oracle allows the model to learn tabular reasoning while protecting the table data.

Source

Training natural language processing (NLP) models using knowledge graphs to improve their performance. The method involves processing a text sample based on triples extracted from the knowledge graph. This fused knowledge representation is used to train the NLP model. The triples contain entity and relation information in addition to the text. The fused knowledge vector contains features like entity meanings and relation contexts not present in the original text. This improves NLP model accuracy compared to just the text input.

Source

Question answering (QA) system based on a complex reasoning pipeline to improve performance on complex questions like bridge-type multi-hop QA. The system recognizes the reasoning type of an input question, decomposes it into sub-questions, and performs optimized reasoning pipelines for each type. The pipelines involve separate modules for question restructuring, search, and document reading. This allows customized reasoning for different question types instead of relying on a single large language model.

Source

Unsupervised multi-hop search for finding relationships between entities in a document corpus. The search involves iteratively finding intermediate entities that connect the initial entities. It uses graph search techniques like A* and beam search to efficiently find relationships through multiple hops. The search starts with a user-selected pair of entities, finds documents containing both, then searches for an intermediary entity in those documents. If too few documents are found, it repeats with more intermediate entities. By selectively expanding promising paths, it avoids exponential complexity.

Source

This paper addresses key challenges in knowledge injection for large language models, including static representation, difficulty updating, and limited domain adaptability. It proposes a dynamic fine-tuning framework designed injection. The is based on parameter-efficient tuning strategies introduces learnable adapter modules gating mechanisms. These components enable selective integration control of both structured unstructured external knowledge. In the model design, query encoder extracts semantic vectors from input text. are matched with an base to construct subset. subset guides task generation. Adapter units then applied across layers adjust enhancement. ensures that contributes reasoning while preserving model's original capabilities. A unified joint loss function also introduced. coordinates optimization between modeling alignment objectives. To evaluate proposed method, WikiHop dataset multi-hop question answering used. behavior analyzed under various experimental settings, different parameter update ratios, densities, cross-domain transfer scenarios. results show method ach... Read More

Source

Large Language Models (LLMs) often tend to hallucinate, especially on domain-specific tasks and that require reasoning. Previously, we introduced SubGraph Retrieval Augmented Generation (SG-RAG) as a novel GraphRAG method for multi-hop question answering. SG-RAG leverages Cypher queries search the given knowledge graph retrieve necessary subgraph answer question. The results from our previous work showed higher performance of compared traditional (RAG). In this work, further enhance by proposing an additional step called Merging Ordering Triplets (MOT). new MOT seeks decrease redundancy in retrieved triplets applying hierarchical merging subgraphs. Moreover, it provides ordering among using Breadth First Search (BFS) traversal algorithm. We conducted experiments MetaQA benchmark, which is proposed question-answering movies domain. Our show more accurate answers than Chain-of-Though Graph Chain-of-Though. also find out (up some point) highly overlapping subgraphs defining order helps LLM generate precise answers.

Source

Retrieval-Augmented Generation (RAG) systems have shown significant potential for domain-specific Question Answering (QA) tasks, although persistent challenges in retrieval precision and context selection continue to hinder their effectiveness. This study introduces Document Graph RAG (GraphRAG), a novel framework that bolsters robustness enhances answer generation by incorporating Knowledge Graphs (KGs) built upon documents intrinsic structure into the pipeline. Through application of Design Science Research methodology, we systematically design, implement, evaluate GraphRAG, leveraging graph-based document structuring keyword-based semantic linking mechanism improve quality. The evaluation, conducted on well-established datasets including SQuAD, HotpotQA, newly developed manufacturing dataset, demonstrates consistent performance gains over naive baseline across both metrics. results indicate GraphRAG improves Context Relevance metrics, with task-dependent optimizations chunk size, keyword density, top-k further enhancing performance. Notably, multi-hop questions benefit most fro... Read More

Source

A system that uses generative AI to intelligently query heterogeneous data sources to answer natural language questions. The system breaks down complex questions into simpler sub-questions and uses AI to identify the appropriate data source for each sub-question. It then generates custom queries for each data source and executes them to gather the necessary information. The AI summarizes the results and returns a coherent response to the original question. The AI avoids sharing or training on the actual data, preserving privacy while leveraging AI's query generation and data source selection abilities.

Source

Using a retrieval system in conjunction with large language models (LLMs) to improve accuracy and reduce the need for re-training when underlying information changes. The system retrieves relevant documents based on a user's query using a retriever. These documents are then combined with the original query to form a prompt for the LLM. The LLM generates an output using this augmented prompt that incorporates the retrieved documents' information. This provides a more accurate response compared to just using the LLM alone, especially for complex domains where the LLM may lack specific knowledge.

Source

Creating a dynamic workflow data structure that can tailor responses to specific user queries and contexts. The method involves using a large language model (LLM) to process a corpus of documents, classify user queries based on context, construct a workflow structure from the classified segments, and generate personalized responses. The LLM segments the documents, analyzes the segments' semantics, and organizes them into a workflow structure. This structure is then used to generate customized responses for user queries based on the segment classification and query context.

Source

Using knowledge graphs and vector searches to improve the efficiency and accuracy of AI question answering systems. The method involves decomposing user queries into tasks, searching a knowledge graph to identify relevant entities, searching a vector database using the entities to find text chunks, and using an LLM to generate answers based on the identified text. This refined search scope improves the LLM's efficiency and accuracy compared to searching large text databases directly.

Source

Automatically generating context-based responses to natural language queries using knowledge graphs in combination with artificial intelligence techniques. The system generates queries in a predetermined query language from natural language inputs, processes them using enterprise knowledge graphs and AI models, and combines the results to provide more accurate and comprehensive responses. This allows users without technical knowledge to search and query knowledge graphs using natural language and make decisions based on the responses.

Source

Multi-Hop Question Answering over Knowledge Graphs (MHQA-KG) plays a pivotal role in various applications, including but not limited to Question Answering, Recommendation Systems, and Semantic Search. Nevertheless, current models for MHQA have limitations in their ability to grasp all the information included in the question, resulting a reduction in accuracy when producing answers. In order to mitigate this limitation, this paper proposes a novel Multi-Hop Question Answering over Knowledge Graphs approach. It mainly utilizes question and path embedding to answer multi-hop questions, significantly improving accuracy. This approach effectively captures auxiliary information that may present in the question. The experimental findings provide evidence that the suggested methodology outperforms the current state-of-the-art models, achieving highly accurate outcomes with improvements.

Source

Large language models (LLMs) have recently demonstrated remarkable performance across various Natual Language Processing tasks. In the field of multi-hop reasoning, the Chain-of-thought (CoT) prompt method has emerged as a paradigm, using curated stepwise reasoning demonstrations to enhance LLM's ability to reason and produce coherent rational pathways. To ensure the accuracy, reliability, and traceability of the generated answers, many studies have incorporated information retrieval (IR) to provide LLMs with external knowledge. However, existing CoT with IR methods decomposes questions into sub-questions based on a single compositionality type, which limits their effectiveness for questions involving multiple compositionality types. Additionally, these methods suffer from inefficient retrieval, as complex questions often contain abundant information, leading to the retrieval of irrelevant information inconsistent with the query's intent. In this work, we propose a novel question decomposition framework called TRQA for multi-hop question answering, which addresses these limitations. ... Read More

Source

Multi-hop Knowledge Graph Question Answering aims at finding an entity to answer natural language questions from knowledge graphs. When humans perform multi-hop reasoning, people tend to focus on specific relations across different hops and confirm the next entity. Therefore, most algorithms choose the wrong specific relation, which makes the system deviate from the correct reasoning path. The specific relation at each hop plays an important role in multi-hop question answering. Existing work mainly relies on the question representation as relation information, which cannot accurately calculate the specific relation distribution. In this article, we propose an interpretable assistance framework that fully utilizes the relation embeddings to assist in calculating relation distributions at each hop. Moreover, we employ the fusion attention mechanism to ensure the integrity of relation information and hence to enrich the relation embeddings. The experimental results on three English datasets and one Chinese dataset demonstrate that our method significantly outperforms all baselines. The... Read More

Source

Multi-Hop Question Answering (MHQA) tasks present a significant challenge for large language models (LLMs) due to the intensive knowledge required. Current solutions, like Retrieval-Augmented Generation, typically retrieve potential documents from an external corpus to read an answer. However, the performance of this retrieve-then-read paradigm is constrained by the retriever and the inevitable noise in the retrieved documents. To mitigate these challenges, we introduce a novel generate-then-ground (GenGround) framework, synergizing the parametric knowledge of LLMs and external documents to solve a multi-hop question. GenGround empowers LLMs to alternate two phases until the final answer is derived: (1) formulate a simpler, single-hop question and directly generate the answer; (2) ground the question-answer pair in retrieved documents, amending any wrong predictions in the answer. We also propose an instructional grounding distillation method to generalize our method into smaller models. Extensive experiments conducted on four datasets illustrate the superiority of our method.

Source

<title>Abstract</title> Multi-hop knowledge graph question answering aims to find answer entities from the knowledge graph based on natural language questions. This is a challenging task as it requires precise reasoning about entity relationships at each step. When humans perform multi-hop reasoning, they usually focus on specific relations between different hops and determine the next entity. However, most algorithms often choose the wrong specific relations, causing the system to deviate from the correct reasoning path. In multi-hop question answering, the specific relation between each hop is crucial. The existing TransferNet model mainly relies on question representation for relational reasoning, but cannot accurately calculate the specific relational distribution, which will profoundly affect question answering performance. On this basis, this paper proposes an interpretable assiatance framework, which makes full use of relation embedding and question semantics, and uses the attention mechanism to cross-fuse the relevant information of them to assist in calculating the relation ... Read More

Source

Multi-hop QA (MHQA) involves step-by-step reasoning to answer complex questions and find multiple relevant supporting facts. However, Existing large language models'(LLMs) reasoning ability in multi-hop question answering remains exploration, which is inadequate in answering multi-hop questions. Moreover, it is unclear whether LLMs follow a desired reasoning chain to reach the right final answer. In this paper, we propose a \textbf{gen}erative question \textbf{dec}omposition method (GenDec) from the perspective of explainable QA by generating independent and complete sub-questions based on incorporating additional extracted evidence for enhancing LLMs' reasoning ability in RAG. To demonstrate the impact, generalization, and robustness of Gendec, we conduct two experiments, the first is combining GenDec with small QA systems on paragraph retrieval and QA tasks. We secondly examine the reasoning capabilities of various state-of-the-art LLMs including GPT-4 and GPT-3.5 combined with GenDec. We experiment on the HotpotQA, 2WikihopMultiHopQA, MuSiQue, and PokeMQA datasets.

Source

Multi-hop query answering over a Knowledge Graph (KG) involves traversing one or more hops from the start node to answer a query. Path-based and logic-based methods are state-of-the-art for multi-hop question answering. The former is used in link prediction tasks. The latter is for answering complex logical queries. The logical multi-hop querying technique embeds the KG and queries in the same embedding space. The existing work incorporates First Order Logic (FOL) operators, such as conjunction ($\wedge$), disjunction ($\vee$), and negation ($\neg$), in queries. Though current models have most of the building blocks to execute the FOL queries, they cannot use the dense information of multi-modal entities in the case of Multi-Modal Knowledge Graphs (MMKGs). We propose RConE, an embedding method to capture the multi-modal information needed to answer a query. The model first shortlists candidate (multi-modal) entities containing the answer. It then finds the solution (sub-entities) within those entities. Several existing works tackle path-based question-answering in MMKGs. However, to ... Read More

Source

Large Language Models (LLMs) have shown proficiency in question-answering tasks but often struggle to integrate real-time knowledge updates, 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 tailored 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 na\"ive similarity-based searches might miss. Additionally, our framework incorporates 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... Read More

Source

Multi-hop question answering (MHQA) requires a model to retrieve and integrate information from multiple passages to answer a complex question. Recent systems leverage the power of large language models and integrate evidence retrieval with reasoning prompts (e.g., chain-of-thought reasoning) for the MHQA task. However, the complexities in the question types (bridge v.s. comparison questions) and the reasoning types (sequential v.s. parallel reasonings) require more novel and fine-grained prompting methods to enhance the performance of MHQA under the zero-shot setting. In this paper, we propose STOC-TOT, a stochastic tree-of-thought reasoning prompting method with constrained decoding for MHQA and conduct a detailed comparison with other reasoning prompts on different question types and reasoning types. Specifically, we construct a tree-like reasoning structure by prompting the model to break down the original question into smaller sub-questions to form different reasoning paths. In addition, we prompt the model to provide a probability estimation for each reasoning path at each reas... Read More

Source

With the advent of the information age, it will be more troublesome to search for a lot of relevant knowledge to find the information you need. Text reasoning is a very basic and important part of multi-hop question and answer tasks. This paper aims to study the integrity, uniformity, and speed of computational intelligence inference data capabilities. That is why multi-hop reasoning came into being, but it is still in its infancy, that is, it is far from enough to conduct multi-hop question and answer questions, such as search breadth, process complexity, response speed, comprehensiveness of information, etc. This paper makes a text comparison between traditional information retrieval and computational intelligence through corpus relevancy and other computing methods. The study finds that in the face of multi-hop question and answer reasoning, the reasoning data that traditional retrieval methods lagged behind in intelligence are about 35% worse. It shows that computational intelligence would be more complete, unified, and faster than traditional retrieval methods. This paper also i... Read More

Source

Retrieval-augmented generation (RAG) augments large language models (LLM) by retrieving relevant knowledge, showing promising potential in mitigating LLM hallucinations and enhancing response quality, thereby facilitating the great adoption of LLMs in practice. However, we find that existing RAG systems are inadequate in answering multi-hop queries, which require retrieving and reasoning over multiple pieces of supporting evidence. Furthermore, to our knowledge, no existing RAG benchmarking dataset focuses on multi-hop queries. In this paper, we develop a novel dataset, MultiHop-RAG, which consists of a knowledge base, a large collection of multi-hop queries, their ground-truth answers, and the associated supporting evidence. We detail the procedure of building the dataset, utilizing an English news article dataset as the underlying RAG knowledge base. We demonstrate the benchmarking utility of MultiHop-RAG in two experiments. The first experiment compares different embedding models for retrieving evidence for multi-hop queries. In the second experiment, we examine the capabilities o... Read More

Source

Large Language Models (LLMs) have demonstrated significant capabilities across numerous application domains. A key challenge is to keep these models updated with latest available information, which limits the true potential of these models for the end-applications. Although, there have been numerous attempts for LLMs Knowledge Editing (KE), i.e., to edit the LLMs prior knowledge and in turn test it via Multi-hop Question Answering (MQA), yet so far these studies are primarily focused on English language. To bridge this gap, in this paper we propose: Multi-hop Questioning Answering under Knowledge Editing for Arabic Language (MQA-KEAL). MQA-KEAL stores knowledge edits as structured knowledge units in the external memory. In order to solve multi-hop question, it first uses task-decomposition to decompose the question into smaller sub-problems. Later for each sub-problem, it iteratively queries the external memory and/or target LLM in order to generate the final response. In addition, we also contribute MQUAKE-AR (Arabic translation of English benchmark MQUAKE), as well as a new benchma... Read More

Source

Abstract Question-answering systems require retrieving evidence from multiple documents or paragraphs and reasoning over them to meet users' information needs and answer their complex questions. On the other hand, the Explainability and comprehensibility of the predictions made by question-answering systems pose a challenge. In this paper, a content-based reasoning approach based on graph-based machine reading comprehension methods is proposed to answer multi-hop complex questions. In this approach, relevant paragraphs are selected in a two-step process after receiving the input of a multi-hop complex question. Then, to facilitate content-based reasoning and utilize the evidence related to the multi-hop complex question in the retrieved paragraphs, an incoherent graph infrastructure is constructed. Subsequently, a graph neural network and a transformer are employed as an encoder to extract the content-based answer relevant to the question from the graph infrastructure. Finally, to overcome the challenge of interpretability in the question-answering system, a transformer and the predi... Read More

Source

Open Domain Multi-Hop Question Answering (ODMHQA) plays a crucial role in Natural Language Processing (NLP) by aiming to answer complex questions through multi-step reasoning over retrieved information from external knowledge sources. Recently, Large Language Models (LLMs) have demonstrated remarkable performance in solving ODMHQA owing to their capabilities including planning, reasoning, and utilizing tools. However, LLMs may generate off-topic answers when attempting to solve ODMHQA, namely the generated answers are irrelevant to the original questions. This issue of off-topic answers accounts for approximately one-third of incorrect answers, yet remains underexplored despite its significance. To alleviate this issue, we propose the Discriminate->Re-Compose->Re- Solve->Re-Decompose (Dr3) mechanism. Specifically, the Discriminator leverages the intrinsic capabilities of LLMs to judge whether the generated answers are off-topic. In cases where an off-topic answer is detected, the Corrector performs step-wise revisions along the reversed reasoning chain (Re-Compose->Re-Solve->Re-Decom... Read More

Source

Visual Question Answering (VQA) often involves diverse reasoning scenarios across Vision and Language (V&L). Most prior VQA studies, however, have merely focused on assessing the model's overall accuracy without evaluating it on different reasoning cases. Furthermore, some recent works observe that conventional Chain-of-Thought (CoT) prompting fails to generate effective reasoning for VQA, especially for complex scenarios requiring multi-hop reasoning. In this paper, we propose II-MMR, a novel idea to identify and improve multi-modal multi-hop reasoning in VQA. In specific, II-MMR takes a VQA question with an image and finds a reasoning path to reach its answer using two novel language promptings: (i) answer prediction-guided CoT prompt, or (ii) knowledge triplet-guided prompt. II-MMR then analyzes this path to identify different reasoning cases in current VQA benchmarks by estimating how many hops and what types (i.e., visual or beyond-visual) of reasoning are required to answer the question. On popular benchmarks including GQA and A-OKVQA, II-MMR observes that most of their VQA que... Read More

Source

With the continuous increase of Internet information, the traditional single-hop query is not enough to meet the needs of users. For complex problems, multi-hop KGQA requires reasoning at multiple edges of the KG to arrive at the correct answer. KGS often lack many links, which poses additional challenges for KGQAs especially multi-jump KGQAs. In this paper, the main multi-hop question answering algorithms are divided into two categories: embedded-based multi-hop knowledge question answering reasoning and linked multi-hop knowledge question answering reasoning. The results show that the EmbedKGQA model performs better in prediction reasoning under the knowledge map with missing links by analyzing and comparing the performance of the subgraph matching and embedding prediction models on the Meta and WebQestionSP data sets. Finally, in view of the absence of knowledge graph data in practical application, we propose the development prospect of knowledge graph multi-hop algorithm from the two directions of combining pre-training with knowledge graph and using multi-modal model to expand t... Read More

Source

Knowledge base multi-hop question answering must perform hop-by-hop reasoning on long paths to obtain answer entities. However, due to the lack of datasets with correct path annotations, the model's training can only rely on the results of the last hop answer for updating. A more innovative approach is to apply the knowledge distillation method to the question-answering task, and the complex teacher model learns the entity distribution of each hop as the intermediate supervision signal of the student model. However, the structure of this dual model needs to be simplified. At the same time, in the backward reasoning process from the candidate answer entity to the topic entity, this method only uses the last hop result of the forward as the initialization, which does not guarantee that the backward reasoning process can conform to the semantic information of the question sentence. To address these challenges, we design a structurally simple model capable of constraining the intermediate reasoning process through forward and backward bidirectional reasoning. Moreover, we updated the ini... Read More

Source

Abstract Multi-hop path query answering is a complex task in which a path needs to be inferred from a knowledge graph that contains a head entity and multi-hop relations. The objective is to identify the corresponding tail entity accurately. The representation of the path is a critical factor in this task. However, existing methods do not adequately consider the context information of the entities and relations in the path. To address this issue, this paper proposes a novel multi-hop path query answering model that utilizes an enhanced reasoning path feature representation to incorporate intermediate entity information and improve the accuracy of path query answering. The proposed model utilizes the neighborhood to aggregate the entity representation in the reasoning path. It employs a recurrent skipping network to splice the embedding of the relationship and the entity in the reasoning path based on their weight. Additionally, the model adds the position representation to obtain the reasoning path representation. Moreover, the model uses Bi-GRU and Transformer to obtain the local an... Read More

Source

Question generation is the parallel task of question answering, where given an input context and, optionally, an answer, the goal is to generate a relevant and fluent natural language question. Although recent works on question generation have experienced success by utilizing sequence-to-sequence models, there is a need for question generation models to handle increasingly complex input contexts to produce increasingly detailed questions. Multi-hop question generation is a more challenging task that aims to generate questions by connecting multiple facts from multiple input contexts. In this work, we apply a transformer model to the task of multi-hop question generation without utilizing any sentence-level supporting fact information. We utilize concepts that have proven effective in single-hop question generation, including a copy mechanism and placeholder tokens. We evaluate our models performance on the HotpotQA dataset using automated evaluation metrics, including BLEU, ROUGE and METEOR and show an improvement over the previous work.

Source

In this paper, we present the Self-Adapting Reasoning Model (SAR) for solving multi-hop question answering (MHQA) tasks, where the QA system is supposed to find the correct answer within the given multiple documents and a multi-hop question. One feasible track on MHQA is question decomposition, based on the idea that a multi-hop question is usually made from several single-hop questions, which are much easier to answer. However, ignoring the inner connection between sub-questions, existing works usually train additional single-hop question-answering models and answer sub-questions separately. To tackle this problem, we design an end-to-end self-adaptive multi-hop reasoning model. Specifically, given a multi-hop question, a question decomposer first decomposes it into two simple questions and identifies the question type. Then, based on the question type, different reasoning strategies are applied for reasoning. This enables our model to be self-adapting and more explainable regarding different types of questions. Experiments are carried out to demonstrate the effectiveness of our mod... Read More

Source

Multi-hop Question Answering has recently received particular attention in research and practice, especially in the context of conversational systems and answering complex questions. Various architectures have been proposed to answer these complex multi-hop questions. However, in real-world scenarios, a conversational system should answer both simple (single-hop) and complex (multi-hop) questions. In this work, we propose an efficient BERT question classifier that supports retrievers in the question-answering systems to process single- and multi-hop questions. We also released a mixed dataset consisting of both single- and multi-hop questions. We show that utilizing our classifier inside the QA system can improve these systems' accuracy and enable them to answer both kinds of questions considering their complexity.

Source

Question generation is the parallel task of question answering, where given an input context and optionally, an answer, the goal is to generate a relevant and fluent natural language question. Although recent works on question generation have experienced success by utilizing sequence-to-sequence models, there is a need for question generation models to handle increasingly complex input contexts with the goal of producing increasingly elaborate questions. Multi-hop question generation is a more challenging task that aims to generate questions by connecting multiple facts from multiple input contexts. In this work we apply a transformer model to the task of multi-hop question generation, without utilizing any sentence-level supporting fact information. We utilize concepts that have proven effective in single-hop question generation, including a copy mechanism and placeholder tokens. We evaluate our model's performance on the HotpotQA dataset using automated evaluation metrics and human evaluation, and show an improvement over the previous works.&#x0D;

Source

The task of multi-hop question generation (QG) seeks to generate questions that require a complex reasoning process that spans multiple sentences and answers. Beyond the conventional challenges of what to ask and how to ask, multi-hop QG necessitates sophisticated reasoning from dispersed evidence across multiple sentences. To address these challenges, a knowledge graph-enhanced language model (KGEL) has been developed to imitate human reasoning for multi-hop questions.The initial step in KGEL involves encoding the input sentence with a pre-trained GPT-2 language model to obtain a comprehensive semantic context representation. Next, a knowledge graph is constructed using the entities identified within the context. The critical information in the graph that is related to the answer is then utilized to update the context representations through an answer-aware graph attention network (GAT). Finally, the multi-head attention generation module (MHAG) is performed over the updated latent representations of the context to generate coherent questions. Human evaluations demonstrate that KGEL... Read More

Source

A question answering system aims to answer the asked question with relevant responses thus sufficing the re-quested query asked in natural language by responding in the same language. Knowledge Graph Question Answering (KGQA) aims to answer questions asked by the user on a paragraph from a knowledge graph (KG). A strongly connected KG is essential in picking out answers for the requested question. This is because the KG is traversed to select the answer. A well connected KG thus provides a relevant answer. The knowledge graph is built by identifying the subject, the object and the relation for every sentence in the input text or knowledge base. Questions are processed to identify the source-relation-target triples which are then matched with that of the triples forming the KG. The challenge is in extracting the entities and relations between them to create the KG. The model's performance is directly proportional to the strength of the KG. Hence, the presence of a well connected KG provides great accuracy while a poorly connected one would break the system. The proposed model is teste... Read More

Source

Multi-hop question answering (QA) involves finding multiple relevant passages and step-by-step reasoning to answer complex questions, indicating a retrieve-and-read paradigm. However, previous retrievers were customized for two-hop questions, and most of them were trained separately across different hops, resulting in a lack of supervision over the entire multi-hop retrieval process and leading to poor performance in complicated scenarios beyond two hops. In this work, we introduce Beam Retrieval, an end-to-end beam retrieval framework for multi-hop QA. This approach models the multi-hop retrieval process in an end-to-end manner by jointly optimizing an encoder and two classification heads across all hops. Moreover, Beam Retrieval maintains multiple partial hypotheses of relevant passages at each step, expanding the search space and reducing the risk of missing relevant passages. To establish a complete QA system, we incorporate a supervised reader or a large language model (LLM). Experimental results demonstrate that Beam Retrieval achieves a nearly 50% improvement compared with bas... Read More

Source

Community question answering aims to connect queries and answers based on users' community behaviors, find the most relevant solutions for newly raised questions, and improve community activity. Existing research mainly focuses on the single-hop method, which relies on a particular feature for simple question answering. However, multi-hop question answering, which depends on multi-step reasoning to solve complex questions, still faces the destitute adaptability caused by various entity features and complex entity correlations. In this work, we put forward a Multi-Hop Community Question-Answering method called MHCQA, that combines the multi-aspect features of community entities, constrains the answer retrieval process from different perspectives and improves the adaptability of multi-hop answer retrieval. Firstly, we utilize phrases for representing the semantic features of question-answering and fuse them with entity constitution properties through heterogeneous graphs to effectively indicate the diverse relationships among entities in communities. At the same time, we bring a dynami... Read More

Source

Modern systems for multi-hop question answering (QA) typically break questions into a sequence of reasoning steps, termed chain-of-thought (CoT), before arriving at a final answer. Often, multiple chains are sampled and aggregated through a voting mechanism over the final answers, but the intermediate steps themselves are discarded. While such approaches improve performance, they do not consider the relations between intermediate steps across chains and do not provide a unified explanation for the predicted answer. We introduce Multi-Chain Reasoning (MCR), an approach which prompts large language models to meta-reason over multiple chains of thought, rather than aggregate their answers. MCR examines different reasoning chains, mixes information between them and selects the most relevant facts in generating an explanation and predicting the answer. MCR outperforms strong baselines on 7 multi-hop QA datasets. Moreover, our analysis reveals that MCR explanations exhibit high quality, enabling humans to verify its answers.

Source

Explainable multi-hop question answering (QA) not only predicts answers but also identifies rationales, i. e. subsets of input sentences used to derive the answers. This problem has been extensively studied under the supervised setting, where both answer and rationale annotations are given. Because rationale annotations are expensive to collect and not always available, recent efforts have been devoted to developing methods that do not rely on supervision for rationales. However, such methods have limited capacities in modeling interactions between sentences, let alone reasoning across multiple documents. This work proposes a principled, probabilistic approach for training explainable multi-hop QA systems without rationale supervision. Our approach performs multi-hop reasoning by explicitly modeling rationales as sets, enabling the model to capture interactions between documents and sentences within a document. Experimental results show that our approach is more accurate at selecting rationales than the previous methods, while maintaining similar accuracy in predicting answers.

Source

Despite readily memorizing world knowledge about entities, pre-trained language models (LMs) struggle to compose together two or more facts to perform multi-hop reasoning in question-answering tasks. In this work, we propose techniques that improve upon this limitation by relying on random-walks over structured knowledge graphs. Specifically, we use soft-prompts to guide LMs to chain together their encoded knowledge by learning to map multi-hop questions to random-walk paths that lead to the answer. Applying our methods on two T5 LMs shows substantial improvements over standard tuning approaches in answering questions that require multi-hop reasoning.

Source