arXiv subjects

Evangelos Kanoulas

Publications and source records attributed to Evangelos Kanoulas.

At least 73 records · Page 4Linked to original sources

arXiv2022

Zero-shot Query Contextualization for Conversational Search

Current conversational passage retrieval systems cast conversational search into ad-hoc search by using an intermediate query resolution step that places the user's question in context of the conversation. While the proposed methods have proven effective, they still assume the availability of large-scale question resolution and conversational search datasets. To waive the dependency on the availability of such data, we adapt a pre-trained token-level dense retriever on ad-hoc search data to perform conversational search with no additional fine-tuning. The proposed method allows to contextualize the user question within the conversation history, but restrict the matching only between question and potential answer. Our experiments demonstrate the effectiveness of the proposed approach. We also perform an analysis that provides insights of how contextualization works in the latent space, in essence introducing a bias towards salient terms from the conversation.

cs.IR↗

arXiv2022

Parameter-Efficient Abstractive Question Answering over Tables or Text

A long-term ambition of information seeking QA systems is to reason over multi-modal contexts and generate natural answers to user queries. Today, memory intensive pre-trained language models are adapted to downstream tasks such as QA by fine-tuning the model on QA data in a specific modality like unstructured text or structured tables. To avoid training such memory-hungry models while utilizing a uniform architecture for each modality, parameter-efficient adapters add and train small task-specific bottle-neck layers between transformer layers. In this work, we study parameter-efficient abstractive QA in encoder-decoder models over structured tabular data and unstructured textual data using only 1.5% additional parameters for each modality. We also ablate over adapter layers in both encoder and decoder modules to study the efficiency-performance trade-off and demonstrate that reducing additional trainable parameters down to 0.7%-1.0% leads to comparable results. Our models out-perform current state-of-the-art models on tabular QA datasets such as Tablesum and FeTaQA, and achieve comparable performance on a textual QA dataset such as NarrativeQA using significantly less trainable parameters than fine-tuning.

cs.CL↗

arXiv2022

Towards Building Economic Models of Conversational Search

Various conceptual and descriptive models of conversational search have been proposed in the literature -- while useful, they do not provide insights into how interaction between the agent and user would change in response to the costs and benefits of the different interactions. In this paper, we develop two economic models of conversational search based on patterns previously observed during conversational search sessions, which we refer to as: Feedback First where the agent asks clarifying questions then presents results, and Feedback After where the agent presents results, and then asks follow up questions. Our models show that the amount of feedback given/requested depends on its efficiency at improving the initial or subsequent query and the relative cost of providing said feedback. This theoretical framework for conversational search provides a number of insights that can be used to guide and inform the development of conversational search agents. However, empirical work is needed to estimate the parameters in order to make predictions specific to a given conversational search setting.

cs.IR↗

arXiv2021

Robustness Evaluation of Entity Disambiguation Using Prior Probes:the Case of Entity Overshadowing

Entity disambiguation (ED) is the last step of entity linking (EL), when candidate entities are reranked according to the context they appear in. All datasets for training and evaluating models for EL consist of convenience samples, such as news articles and tweets, that propagate the prior probability bias of the entity distribution towards more frequently occurring entities. It was previously shown that the performance of the EL systems on such datasets is overestimated since it is possible to obtain higher accuracy scores by merely learning the prior. To provide a more adequate evaluation benchmark, we introduce the ShadowLink dataset, which includes 16K short text snippets annotated with entity mentions. We evaluate and report the performance of popular EL systems on the ShadowLink benchmark. The results show a considerable difference in accuracy between more and less common entities for all of the EL systems under evaluation, demonstrating the effects of prior probability bias and entity overshadowing.

cs.CL↗

arXiv2021

Controllable Recommenders using Deep Generative Models and Disentanglement

In this paper, we consider controllability as a means to satisfy dynamic preferences of users, enabling them to control recommendations such that their current preference is met. While deep models have shown improved performance for collaborative filtering, they are generally not amenable to fine grained control by a user, leading to the development of methods like deep language critiquing. We propose an alternate view, where instead of keyphrase based critiques, a user is provided 'knobs' in a disentangled latent space, with each knob corresponding to an item aspect. Disentanglement here refers to a latent space where generative factors (here, a preference towards an item category like genre) are captured independently in their respective dimensions, thereby enabling predictable manipulations, otherwise not possible in an entangled space. We propose using a (semi-)supervised disentanglement objective for this purpose, as well as multiple metrics to evaluate the controllability and the degree of personalization of controlled recommendations. We show that by updating the disentangled latent space based on user feedback, and by exploiting the generative nature of the recommender, controlled and personalized recommendations can be produced. Through experiments on two widely used collaborative filtering datasets, we demonstrate that a controllable recommender can be trained with a slight reduction in recommender performance, provided enough supervision is provided. The recommendations produced by these models appear to both conform to a user's current preference and remain personalized.

cs.IR↗

arXiv2021

Paint4Poem: A Dataset for Artistic Visualization of Classical Chinese Poems

In this work we propose a new task: artistic visualization of classical Chinese poems, where the goal is to generatepaintings of a certain artistic style for classical Chinese poems. For this purpose, we construct a new dataset called Paint4Poem. Thefirst part of Paint4Poem consists of 301 high-quality poem-painting pairs collected manually from an influential modern Chinese artistFeng Zikai. As its small scale poses challenges for effectively training poem-to-painting generation models, we introduce the secondpart of Paint4Poem, which consists of 3,648 caption-painting pairs collected manually from Feng Zikai's paintings and 89,204 poem-painting pairs collected automatically from the web. We expect the former to help learning the artist painting style as it containshis most paintings, and the latter to help learning the semantic relevance between poems and paintings. Further, we analyze Paint4Poem regarding poem diversity, painting style, and the semantic relevance between poems and paintings. We create abenchmark for Paint4Poem: we train two representative text-to-image generation models: AttnGAN and MirrorGAN, and evaluate theirperformance regarding painting pictorial quality, painting stylistic relevance, and semantic relevance between poems and paintings.The results indicate that the models are able to generate paintings that have good pictorial quality and mimic Feng Zikai's style, but thereflection of poem semantics is limited. The dataset also poses many interesting research directions on this task, including transferlearning, few-shot learning, text-to-image generation for low-resource data etc. The dataset is publicly available.(https://github.com/paint4poem/paint4poem)

cs.CV↗

arXiv2021

Combining Lexical and Dense Retrieval for Computationally Efficient Multi-hop Question Answering

In simple open-domain question answering (QA), dense retrieval has become one of the standard approaches for retrieving the relevant passages to infer an answer. Recently, dense retrieval also achieved state-of-the-art results in multi-hop QA, where aggregating information from multiple pieces of information and reasoning over them is required. Despite their success, dense retrieval methods are computationally intensive, requiring multiple GPUs to train. In this work, we introduce a hybrid (lexical and dense) retrieval approach that is highly competitive with the state-of-the-art dense retrieval models, while requiring substantially less computational resources. Additionally, we provide an in-depth evaluation of dense retrieval methods on limited computational resource settings, something that is missing from the current literature.

cs.IR↗

arXiv2021

Analysing Mixed Initiatives and Search Strategies during Conversational Search

Information seeking conversations between users and Conversational Search Agents (CSAs) consist of multiple turns of interaction. While users initiate a search session, ideally a CSA should sometimes take the lead in the conversation by obtaining feedback from the user by offering query suggestions or asking for query clarifications i.e. mixed initiative. This creates the potential for more engaging conversational searches, but substantially increases the complexity of modelling and evaluating such scenarios due to the large interaction space coupled with the trade-offs between the costs and benefits of the different interactions. In this paper, we present a model for conversational search -- from which we instantiate different observed conversational search strategies, where the agent elicits: (i) Feedback-First, or (ii) Feedback-After. Using 49 TREC WebTrack Topics, we performed an analysis comparing how well these different strategies combine with different mixed initiative approaches: (i) Query Suggestions vs. (ii) Query Clarifications. Our analysis reveals that there is no superior or dominant combination, instead it shows that query clarifications are better when asked first, while query suggestions are better when asked after presenting results. We also show that the best strategy and approach depends on the trade-offs between the relative costs between querying and giving feedback, the performance of the initial query, the number of assessments per query, and the total amount of gain required. While this work highlights the complexities and challenges involved in analyzing CSAs, it provides the foundations for evaluating conversational strategies and conversational search agents in batch/offline settings.

cs.IR↗

arXiv2021

Cross-Market Product Recommendation

We study the problem of recommending relevant products to users in relatively resource-scarce markets by leveraging data from similar, richer in resource auxiliary markets. We hypothesize that data from one market can be used to improve performance in another. Only a few studies have been conducted in this area, partly due to the lack of publicly available experimental data. To this end, we collect and release XMarket, a large dataset covering 18 local markets on 16 different product categories, featuring 52.5 million user-item interactions. We introduce and formalize the problem of cross-market product recommendation, i.e., market adaptation. We explore different market-adaptation techniques inspired by state-of-the-art domain-adaptation and meta-learning approaches and propose a novel neural approach for market adaptation, named FOREC. Our model follows a three-step procedure -- pre-training, forking, and fine-tuning -- in order to fully utilize the data from an auxiliary market as well as the target market. We conduct extensive experiments studying the impact of market adaptation on different pairs of markets. Our proposed approach demonstrates robust effectiveness, consistently improving the performance on target markets compared to competitive baselines selected for our analysis. In particular, FOREC improves on average 24% and up to 50% in terms of nDCG@10, compared to the NMF baseline. Our analysis and experiments suggest specific future directions in this research area. We release our data and code for academic purposes.

cs.IR↗

arXiv2021

Legal Search in Case Law and Statute Law

In this work we describe a method to identify document pairwise relevance in the context of a typical legal document collection: limited resources, long queries and long documents. We review the usage of generalized language models, including supervised and unsupervised learning. We observe how our method, while using text summaries, overperforms existing baselines based on full text, and motivate potential improvement directions for future work.

cs.CL↗

arXiv2021

VerbCL: A Dataset of Verbatim Quotes for Highlight Extraction in Case Law

Citing legal opinions is a key part of legal argumentation, an expert task that requires retrieval, extraction and summarization of information from court decisions. The identification of legally salient parts in an opinion for the purpose of citation may be seen as a domain-specific formulation of a highlight extraction or passage retrieval task. As similar tasks in other domains such as web search show significant attention and improvement, progress in the legal domain is hindered by the lack of resources for training and evaluation. This paper presents a new dataset that consists of the citation graph of court opinions, which cite previously published court opinions in support of their arguments. In particular, we focus on the verbatim quotes, i.e., where the text of the original opinion is directly reused. With this approach, we explain the relative importance of different text spans of a court opinion by showcasing their usage in citations, and measuring their contribution to the relations between opinions in the citation graph. We release VerbCL, a large-scale dataset derived from CourtListener and introduce the task of highlight extraction as a single-document summarization task based on the citation graph establishing the first baseline results for this task on the VerbCL dataset.

cs.CL↗

arXiv2021

Few-Shot Electronic Health Record Coding through Graph Contrastive Learning

Electronic health record (EHR) coding is the task of assigning ICD codes to each EHR. Most previous studies either only focus on the frequent ICD codes or treat rare and frequent ICD codes in the same way. These methods perform well on frequent ICD codes but due to the extremely unbalanced distribution of ICD codes, the performance on rare ones is far from satisfactory. We seek to improve the performance for both frequent and rare ICD codes by using a contrastive graph-based EHR coding framework, CoGraph, which re-casts EHR coding as a few-shot learning task. First, we construct a heterogeneous EHR word-entity (HEWE) graph for each EHR, where the words and entities extracted from an EHR serve as nodes and the relations between them serve as edges. Then, CoGraph learns similarities and dissimilarities between HEWE graphs from different ICD codes so that information can be transferred among them. In a few-shot learning scenario, the model only has access to frequent ICD codes during training, which might force it to encode features that are useful for frequent ICD codes only. To mitigate this risk, CoGraph devises two graph contrastive learning schemes, GSCL and GECL, that exploit the HEWE graph structures so as to encode transferable features. GSCL utilizes the intra-correlation of different sub-graphs sampled from HEWE graphs while GECL exploits the inter-correlation among HEWE graphs at different clinical stages. Experiments on the MIMIC-III benchmark dataset show that CoGraph significantly outperforms state-of-the-art methods on EHR coding, not only on frequent ICD codes, but also on rare codes, in terms of several evaluation indicators. On frequent ICD codes, GSCL and GECL improve the classification accuracy and F1 by 1.31% and 0.61%, respectively, and on rare ICD codes CoGraph has more obvious improvements by 2.12% and 2.95%.

cs.AI↗

arXiv2021

A Large-Scale Analysis of Mixed Initiative in Information-Seeking Dialogues for Conversational Search

Conversational search is a relatively young area of research that aims at automating an information-seeking dialogue. In this paper we help to position it with respect to other research areas within conversational Artificial Intelligence (AI) by analysing the structural properties of an information-seeking dialogue. To this end, we perform a large-scale dialogue analysis of more than 150K transcripts from 16 publicly available dialogue datasets. These datasets were collected to inform different dialogue-based tasks including conversational search. We extract different patterns of mixed initiative from these dialogue transcripts and use them to compare dialogues of different types. Moreover, we contrast the patterns found in information-seeking dialogues that are being used for research purposes with the patterns found in virtual reference interviews that were conducted by professional librarians. The insights we provide (1) establish close relations between conversational search and other conversational AI tasks; and (2) uncover limitations of existing conversational datasets to inform future data collection tasks.

cs.IR↗

arXiv2021

Conversations with Search Engines: SERP-based Conversational Response Generation

In this paper, we address the problem of answering complex information needs by conversing conversations with search engines, in the sense that users can express their queries in natural language, and directly receivethe information they need from a short system response in a conversational manner. Recently, there have been some attempts towards a similar goal, e.g., studies on Conversational Agents (CAs) and Conversational Search (CS). However, they either do not address complex information needs, or they are limited to the development of conceptual frameworks and/or laboratory-based user studies. We pursue two goals in this paper: (1) the creation of a suitable dataset, the Search as a Conversation (SaaC) dataset, for the development of pipelines for conversations with search engines, and (2) the development of astate-of-the-art pipeline for conversations with search engines, the Conversations with Search Engines (CaSE), using this dataset. SaaC is built based on a multi-turn conversational search dataset, where we further employ workers from a crowdsourcing platform to summarize each relevant passage into a short, conversational response. CaSE enhances the state-of-the-art by introducing a supporting token identification module and aprior-aware pointer generator, which enables us to generate more accurate responses. We carry out experiments to show that CaSE is able to outperform strong baselines. We also conduct extensive analyses on the SaaC dataset to show where there is room for further improvement beyond CaSE. Finally, we release the SaaC dataset and the code for CaSE and all models used for comparison to facilitate future research on this topic.

cs.IR↗

arXiv2021

Ranking Clarifying Questions Based on Predicted User Engagement

To improve online search results, clarification questions can be used to elucidate the information need of the user. This research aims to predict the user engagement with the clarification pane as an indicator of relevance based on the lexical information: query, question, and answers. Subsequently, the predicted user engagement can be used as a feature to rank the clarification panes. Regression and classification are applied for predicting user engagement and compared to naive heuristic baselines (e.g. mean) on the new MIMICS dataset [20]. An ablation study is carried out using a RankNet model to determine whether the predicted user engagement improves clarification pane ranking performance. The prediction models were able to improve significantly upon the naive baselines, and the predicted user engagement feature significantly improved the RankNet results in terms of NDCG and MRR. This research demonstrates the potential for ranking clarification panes based on lexical information only and can serve as a first neural baseline for future research to improve on. The code is available online.

cs.IR↗

arXiv2021

Category Aware Explainable Conversational Recommendation

Most conversational recommendation approaches are either not explainable, or they require external user's knowledge for explaining or their explanations cannot be applied in real time due to computational limitations. In this work, we present a real time category based conversational recommendation approach, which can provide concise explanations without prior user knowledge being required. We first perform an explainable user model in the form of preferences over the items' categories, and then use the category preferences to recommend items. The user model is performed by applying a BERT-based neural architecture on the conversation. Then, we translate the user model into item recommendation scores using a Feed Forward Network. User preferences during the conversation in our approach are represented by category vectors which are directly interpretable. The experimental results on the real conversational recommendation dataset ReDial demonstrate comparable performance to the state-of-the-art, while our approach is explainable. We also show the potential power of our framework by involving an oracle setting of category preference prediction.

cs.AI↗

arXiv2020

A Benchmark for Lease Contract Review

Extracting entities and other useful information from legal contracts is an important task whose automation can help legal professionals perform contract reviews more efficiently and reduce relevant risks. In this paper, we tackle the problem of detecting two different types of elements that play an important role in a contract review, namely entities and red flags. The latter are terms or sentences that indicate that there is some danger or other potentially problematic situation for one or more of the signing parties. We focus on supporting the review of lease agreements, a contract type that has received little attention in the legal information extraction literature, and we define the types of entities and red flags needed for that task. We release a new benchmark dataset of 179 lease agreement documents that we have manually annotated with the entities and red flags they contain, and which can be used to train and test relevant extraction algorithms. Finally, we release a new language model, called ALeaseBERT, pre-trained on this dataset and fine-tuned for the detection of the aforementioned elements, providing a baseline for further research

cs.IR↗

arXiv2020

Analysing the Effect of Clarifying Questions on Document Ranking in Conversational Search

Recent research on conversational search highlights the importance of mixed-initiative in conversations. To enable mixed-initiative, the system should be able to ask clarifying questions to the user. However, the ability of the underlying ranking models (which support conversational search) to account for these clarifying questions and answers has not been analysed when ranking documents, at large. To this end, we analyse the performance of a lexical ranking model on a conversational search dataset with clarifying questions. We investigate, both quantitatively and qualitatively, how different aspects of clarifying questions and user answers affect the quality of ranking. We argue that there needs to be some fine-grained treatment of the entire conversational round of clarification, based on the explicit feedback which is present in such mixed-initiative settings. Informed by our findings, we introduce a simple heuristic-based lexical baseline, that significantly outperforms the existing naive baselines. Our work aims to enhance our understanding of the challenges present in this particular task and inform the design of more appropriate conversational ranking models.

cs.IR↗