Reinforcement Learning from Reformulations in Conversational Question Answering over Knowledge Graphs

Kaiser, Magdalena; Roy, Rishiraj Saha; Weikum, Gerhard Saha

doi:10.1145/3506805

Cited by 8 publications

(13 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dynamic sub-graph approach was extended by Kaiser et al (2021) with their model, CONQUER. It uses reinforcement learning to select graph traversal actions.…”

Section: Conversational Qa On Knowledge Graphsmentioning

confidence: 99%

Conversational Information Seeking

Zamani¹,

Trippas²,

Dalton³

et al. 2022

Preprint

View full text Add to dashboard Cite

Conversational information seeking (CIS) is concerned with a sequence of interactions between one or more users and an information system. Interactions in CIS are primarily based on natural language dialogue, while they may include other types of interactions, such as click, touch, and body gestures. This monograph provides a thorough overview of CIS definitions, applications, interactions, interfaces, design, implementation, and evaluation. This monograph views CIS applications as including conversational search, conversational question answering, and conversational recommendation. Our aim is to provide an overview of past research related to CIS, introduce the current state-of-the-art in CIS, highlight the challenges still being faced in the community. and suggest future directions.

show abstract

“…The dynamic sub-graph approach was extended by Kaiser et al (2021) with their model, CONQUER. It uses reinforcement learning to select graph traversal actions.…”

Section: Conversational Qa On Knowledge Graphsmentioning

confidence: 99%

Conversational Information Seeking

Zamani¹,

Trippas²,

Dalton³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…One could, for example, provide users with passages or documents, and ask them to create a sequence of questions from there [6,37]. Alternatively, one could also provide annotators with some conversation from a benchmark so far, and request their continuation in some fashion [20]. Large-scale synthetic benchmarks would try to automate this as far as possible using rules and templates [41].…”

Section: The Convmix Benchmarkmentioning

confidence: 99%

“…Prepending history turns. Adding turns from the history to the beginning of the current question is still considered a simple yet tough-to-beat baseline in almost all ConvQA tasks [8,20,33,51], and so we investigate the same here as well. Specifically, we consider four variants: i) add only the initial turn ⟨𝑞 0 , 𝑎 0 ⟩, as it often establishes the topic of the conversation (Prepend init); ii) add only the previous turn ⟨𝑞 𝑖−1 , 𝑎 𝑖−1 ⟩, as it sets immediate context for the current information need (Prepend prev); iii) add both initial and previous turns (Prepend init+prev); and iv) add all turns {⟨𝑞 𝑡 , 𝑎 𝑡 ⟩} 𝑖−1 𝑡 =0 (Prepend all).…”

Section: Baselinesmentioning

confidence: 99%

“…In ConvQA over KBs, the conversational flow is typically captured in a context graph, either explicity [8] using iterative expansion, or implicitly [10,20,26] by maintaining a set of context entities and predicates. Conversational KB history can be incorporated using encodings [10,26,44] or heuristic subsets of the history [20], similar to ConvQA over text. All existing methods consider only one information source for answering conversations, which limits their answer coverage.…”

Section: Related Workmentioning

confidence: 99%

“…State-of-the-art works on ConvQA make use of single information sources: either curated knowledge bases (KB) [8,10,18,20,22,26,29,41,44], or unstructured text collections [5,13,31,33,34], or Web tables [14,27], but only one of these. Questions 𝑞 0 and 𝑞 2 can be answered more conveniently using KBs like Wikidata [53], YAGO [46], or DBpedia [1], that store factual world knowledge in compact RDF triples.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Conversational Question Answering on Heterogeneous Sources

Christmann¹,

Roy²,

Weikum³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Conversational question answering (ConvQA) tackles sequential information needs where contexts in follow-up questions are left implicit. Current ConvQA systems operate over homogeneous sources of information: either a knowledge base (KB), or a text corpus, or a collection of tables. This paper addresses the novel issue of jointly tapping into all of these together, this way boosting answer coverage and confidence. We present Convinse, an end-to-end pipeline for ConvQA over heterogeneous sources, operating in three stages: i) learning an explicit structured representation of an incoming question and its conversational context, ii) harnessing this framelike representation to uniformly capture relevant evidences from KB, text, and tables, and iii) running a fusion-in-decoder model to generate the answer. We construct and release the first benchmark, ConvMix, for ConvQA over heterogeneous sources, comprising 3000 real-user conversations with 16000 questions, along with entity annotations, completed question utterances, and question paraphrases. Experiments demonstrate the viability and advantages of our method, compared to state-of-the-art baselines. CCS CONCEPTS• Information systems → Question answering.

show abstract

Conversational Question Answering over Knowledge Graphs

Kacupaj,

Singh,

Maleshkova

et al. 2024

Event Analytics Across Languages and Communities

View full text Add to dashboard Cite

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

show abstract

Reinforcement Learning from Reformulations in Conversational Question Answering over Knowledge Graphs

Cited by 8 publications

References 0 publications

Conversational Information Seeking

Conversational Information Seeking

Conversational Question Answering on Heterogeneous Sources

Conversational Question Answering over Knowledge Graphs

Contact Info

Product

Resources

About