Search

Conversational search applications offer the prospect of improved user experience in information seeking via agent support. However, it is not clear how searchers will respond to this mode of engagement, in comparison to a conventional user-driven search interface, such as those found in a standard web search engine. We describe a laboratory-based study directly comparing user behaviour for a conventional search interface (CSI) with that of an agent-mediated multiview conversational search interface (MCSI) which extends the CSI. User reaction and search outcomes of the two interfaces are compared using implicit evaluation using five analysis methods: claiming to have a better search experience in contrast to a corresponding standard search interface.

To facilitate effective translation modeling and translation studies, one of the crucial questions to address is how to assess translation quality. From the perspectives of accuracy, reliability, repeatability and cost, translation quality assessment (TQA) itself is a rich and challenging task. In this work, we present a high-level and concise survey of TQA methods, including both manual judgement criteria and automated evaluation metrics, which we classify into further detailed sub-categories. We hope that this work will be an asset for both translation model researchers and quality assessment researchers. In addition, we hope that it will enable practitioners to quickly develop a better understanding of the conventional TQA field, and to find corresponding closely relevant evaluation solutions for their own needs. This work may also serve inspire further development of quality assessment and evaluation methodologies for other natural language processing (NLP) tasks in addition to machine translation (MT), such as automatic text summarization (ATS), natural language understanding (NLU) and natural language generation (NLG)., Accepted to 23rd Nordic Conference on Computational Linguistics (NoDaLiDa 2021): Workshop on Modelling Translation: Translatology in the Digital Age (MoTra21). arXiv admin note: substantial text overlap with arXiv:1605.04515

Conversational search presents opportunities to support users in their search activities to improve the effectiveness and efficiency of search while reducing their cognitive load. Limitations of the potential competency of conversational agents restrict the situations for which conversational search agents can replace human intermediaries. It is thus more interesting, initially at least, to investigate opportunities for conversational interaction to support less complex information retrieval tasks, such as typical web search, which do not require human-level intelligence in the conversational agent. In order to move towards the development of a system to enable conversational search of this type, we need to understand their required capabilities. To progress our understanding of these, we report a study examining the behaviour of users when using a standard web search engine, designed to enable us to identify opportunities to support their search activities using a conversational agent., Accepted in SIGIR 2018 Second International Workshop on Conversational Approaches to Information Retrieval (CAIR 18), July 12, 2018, Ann Arbor Michigan, USA

Chinese character decomposition has been used as a feature to enhance Machine Translation (MT) models, combining radicals into character and word level models. Recent work has investigated ideograph or stroke level embedding. However, questions remain about different decomposition levels of Chinese character representations, radical and strokes, best suited for MT. To investigate the impact of Chinese decomposition embedding in detail, i.e., radical, stroke, and intermediate levels, and how well these decompositions represent the meaning of the original character sequences, we carry out analysis with both automated and human evaluation of MT. Furthermore, we investigate if the combination of decomposed Multiword Expressions (MWEs) can enhance the model learning. MWE integration into MT has seen more than a decade of exploration. However, decomposed MWEs has not previously been explored., Comment: Accepted to publish in NoDaLiDa2021

Conversational search (CS) has recently become a significant focus of the information retrieval (IR) research community. Multiple studies have been conducted which explore the concept of conversational search. Understanding and advancing research in CS requires careful and detailed evaluation. Existing CS studies have been limited to evaluation based on simple user feedback on task completion. We propose a CS evaluation framework which includes multiple dimensions: search experience, knowledge gain, software usability, cognitive load and user experience, based on studies of conversational systems and IR. We introduce these evaluation criteria and propose their use in a framework for the evaluation of CS systems., Comment: Accepted in MICROS (Mixed-Initiative ConveRsatiOnal Systems) Workshop at 43rd European Conference on Information Retrieval

The Podcast Track is new at the Text Retrieval Conference (TREC) in 2020. The podcast track was designed to encourage research into podcasts in the information retrieval and NLP research communities. The track consisted of two shared tasks: segment retrieval and summarization, both based on a dataset of over 100,000 podcast episodes (metadata, audio, and automatic transcripts) which was released concurrently with the track. The track generated considerable interest, attracted hundreds of new registrations to TREC and fifteen teams, mostly disjoint between search and summarization, made final submissions for assessment. Deep learning was the dominant experimental approach for both search experiments and summarization. This paper gives an overview of the tasks and the results of the participants' experiments. The track will return to TREC 2021 with the same two tasks, incorporating slight modifications in response to participant feedback.

The TREC Video Retrieval Evaluation (TRECVID) is a TREC-style video analysis and retrieval evaluation with the goal of promoting progress in research and development of content-based exploitation and retrieval of information from digital video via open, metrics-based evaluation. Over the last twenty years this effort has yielded a better understanding of how systems can effectively accomplish such processing and how one can reliably benchmark their performance. TRECVID has been funded by NIST (National Institute of Standards and Technology) and other US government agencies. In addition, many organizations and individuals worldwide contribute significant time and effort. TRECVID 2020 represented a continuation of four tasks and the addition of two new tasks. In total, 29 teams from various research organizations worldwide completed one or more of the following six tasks: 1. Ad-hoc Video Search (AVS), 2. Instance Search (INS), 3. Disaster Scene Description and Indexing (DSDI), 4. Video to Text Description (VTT), 5. Activities in Extended Video (ActEV), 6. Video Summarization (VSUM). This paper is an introduction to the evaluation framework, tasks, data, and measures used in the evaluation campaign., Comment: TRECVID 2020 Workshop Overview Paper. arXiv admin note: substantial text overlap with arXiv:2009.09984

Live video commenting systems are an emerging feature of online video sites. Recently the Chinese video sharing platform Bilibili, has popularised a novel captioning system where user comments are displayed as streams of moving subtitles overlaid on the video playback screen and broadcast to all viewers in real-time. LiveBot was recently introduced as a novel Automatic Live Video Commenting (ALVC) application. This enables the automatic generation of live video comments from both the existing video stream and existing viewers comments. In seeking to reproduce the baseline results reported in the original Livebot paper, we found differences between the reproduced results using the project codebase and the numbers reported in the paper. Further examination of this situation suggests that this may be caused by a number of small issues in the project code, including a non-obvious overlap between the training and test sets. In this paper, we study these discrepancies in detail and propose an alternative baseline implementation as a reference for other researchers in this field., Comment: 4 pages, 2 figures

A vast amount of audio-visual data is available on the Internet thanks to video streaming services, to which users upload their content. However, there are difficulties in exploiting available data for supervised statistical models due to the lack of labels. Unfortunately, generating labels for such amount of data through human annotation can be expensive, time-consuming and prone to annotation errors. In this paper, we propose a method to automatically extract entity-video frame pairs from a collection of instruction videos by using speech transcriptions and videos. We conduct experiments on image recognition and visual grounding tasks on the automatically constructed entity-video frame dataset of How2. The models will be evaluated on new manually annotated portion of How2 dev5 and val set and on the Flickr30k dataset. This work constitutes a first step towards meta-algorithms capable of automatically construct task-specific training sets.

International audience

A major difficulty in applying word vector embeddings in IR is in devising an effective and efficient strategy for obtaining representations of compound units of text, such as whole documents, (in comparison to the atomic words), for the purpose of indexing and scoring documents. Instead of striving for a suitable method for obtaining a single vector representation of a large document of text, we rather aim for developing a similarity metric that makes use of the similarities between the individual embedded word vectors in a document and a query. More specifically, we represent a document and a query as sets of word vectors, and use a standard notion of similarity measure between these sets, computed as a function of the similarities between each constituent word pair from these sets. We then make use of this similarity measure in combination with standard IR based similarities for document ranking. The results of our initial experimental investigations shows that our proposed method improves MAP by up to $5.77\%$, in comparison to standard text-based language model similarity, on the TREC ad-hoc dataset., Neu-IR '16 SIGIR Workshop on Neural Information Retrieval July 21, 2016, Pisa, Italy

International audience

International audience

This report describes metrics for the evaluation of the effectiveness of segment-based retrieval based on existing binary information retrieval metrics. This metrics are described in the context of a task for the hyperlinking of video segments. This evaluation approach re-uses existing evaluation measures from the standard Cranfield evaluation paradigm. Our adaptation approach can in principle be used with any kind of effectiveness measure that uses binary relevance, and for other segment-baed retrieval tasks. In our video hyperlinking setting, we use precision at a cut-off rank n and mean average precision., Comment: Explanation of evaluation measures for the linking task of the MediaEval Workshop 2013

In this paper we describe our participation in the Interactive Social Book Search task at CLEF 2015. We focus our analysis on differences in search behaviour between native and non-native speakers of English. The analysis is based on both questionnaire and log data. 49 participants out of the 192 total participants are native speakers and the remaining 143 participants are nonnative speakers. In general results show surprisingly few differences in search behaviour between native and non-native speakers. Non-native speakers spent more time on both the focused and the open task than the native speakers, but no significant differences were found in relation to number of queries, query length, depth of results inspection, number of books added to the book-bag, or length of notes explaining why a book was added to the book-bag.

Users looking for books online are confronted with both professional meta-data and user-generated content. The goal of the Interactive Social Book Search Track was to investigate how users used these two sources of information, when looking for books in a leisure context. To this end participants recruited by four teams performed two different tasks using one of two book-search interfaces. Additionally one of the two interfaces also investigated whether user performance can be improved by providing a user-interface that supports multiple search stages.

This paper starts by describing Esfinge, a general domain Portuguese question answering system that uses the redundancy available in the Web as an important resource to find its answers. The paper also presents the strategies employed to participate in CLEF-2004 and discusses the results obtained. Three different strategies were tested: searching the answers only in the CLEF document collection, searching the answers in the Web and using the CLEF document collection to confirm these answers and finally searching the answers only in the Web. The intriguing question of why the system performed better when joining the two information sources, even though it was designed for the Web is discussed; in this connection, different language varieties and some problems of Google are mentioned. The paper concludes describing some of the work planned for the near future.

This paper describes the participation of the XLDB Group in the CLEF monolingual ad hoc task for Portuguese. We present tumba!, a Portuguese search engine and describe its architecture and the underlying assumptions. We discuss the way we used tumba! in CLEF, providing details on our runs and our experiments with ranking algorithms.

A Pilot Question Answering Task has been activated in the Cross-Language Evaluation Forum 2004 with a twofold objective. In the first place, the evaluation of Question Answering systems when they have to answer conjunctive lists, disjunctive lists and questions with temporal restrictions. In the second place, the evaluation of systems' capability to give an accurate self-scoring about the confidence on their answers. In this way, two measures have been designed to be applied on all these different types of questions and to reward systems that give a confidence score with a high correlation with the human assessments. The forty eight runs submitted to the Question Answering Main Track have been taken as a case of study, confirming that some systems are able to give a very accurate score and showing how the measures proposed reward this fact.