1,720,969 research outputs found
Pairing Clustered Inverted Indexes with κ-NN Graphs for Fast Approximate Retrieval over Learned Sparse Representations
Get PDFLearned sparse representations form an effective and interpretable class of embeddings for text retrieval. While exact top-k retrieval over such embeddings faces efficiency challenges, a recent algorithm called Seismic has enabled remarkably fast, highly-accurate approximate retrieval. Seismic statically prunes inverted lists, organizes each list into geometrically-cohesive blocks, and augments each block with a summary vector. At query time, each inverted list associated with a query term is traversed one block at a time in an arbitrary order, with the inner product between the query and summaries determining if a block must be evaluated. When a block is deemed promising, its documents are fully evaluated with a forward index. Seismic is one to two orders of magnitude faster than state-of-the-art inverted index-based solutions and significantly outperforms the winning graph-based submissions to the BigANN 2023 Challenge. In this work, we speed up Seismic further by introducing two innovations to its query processing subroutine. First, we traverse blocks in order of importance, rather than arbitrarily. Second, we take the list of documents retrieved by Seismic and expand it to include the neighbors of each document using an offline k-regular nearest neighbor graph; the expanded list is then ranked to produce the final top-k set. Experiments on two public datasets show that our extension, named SeismicWave, can reach almost-exact accuracy levels and is up to 2.2x faster than Seismic
ChatGPT Versus Modest Large Language Models: An Extensive Study on Benefits and Drawbacks for Conversational Search
Get PDFLarge Language Models (LLMs) are effective in modeling text syntactic and semantic content, making them a strong choice to perform conversational query rewriting. While previous approaches proposed NLP-based custom models, requiring significant engineering effort, our approach is straightforward and conceptually simpler. Not only do we improve effectiveness over the current state-of-the-art, but we also curate the cost and efficiency aspects. We explore the use of pre-trained LLMs fine-tuned to generate quality user query rewrites, aiming to reduce computational costs while maintaining or improving retrieval effectiveness. As a first contribution, we study various prompting approaches - including zero, one, and few-shot methods - with ChatGPT (e.g., gpt-3.5-turbo). We observe an increase in the quality of rewrites leading to improved retrieval. We then fine-tuned smaller open LLMs on the query rewriting task. Our results demonstrate that our fine-tuned models, including the smallest with 780 million parameters, achieve better performance during the retrieval phase than gpt-3.5-turbo. To fine-tune the selected models, we used the QReCC dataset, which is specifically designed for query rewriting tasks. For evaluation, we used the TREC CAsT datasets to assess the retrieval effectiveness of the rewrites of both gpt-3.5-turbo and our fine-tuned models. Our findings show that fine-tuning LLMs on conversational query rewriting datasets can be more effective than relying on generic instruction-tuned models or traditional query reformulation techniques
LongDoc summarization using instruction-tuned large language models for food safety regulations
No full text
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
On Improving Efficiency/Effectiveness trade-offs with Neural Network Compression
No full textDeep Neural Networks (DNNs) deliver state-of-the-art performance in various fields at the price of huge computational requirements.
In this thesis, we propose three solutions to reduce the computational requirements of DNNs in Learning to Rank (LtR), Image Classification, and multi-term Dense Retrieval (DR).
LtR is the field of machine learning employed to rank candidate documents in a search engine. We propose a methodology to train efficient and effective neural networks for LtR by e employing pruning and cross-modal knowledge distillation. Furthermore, we develop analytic time predictors estimating the execution time of sparse and dense neural networks, thus easing the design of neural models matching the desired time requirements.
In Image Classification, we propose Automatic Prune Binarization (APB), a novel compression framework enriching the expressiveness of binary networks with few full-precision weights. Moreover, we design two innovative matrix multiplication algorithms for extremely low bits configurations, based on highly efficient bitwise and logical CPU instructions.
In multi-term DR, we propose two different contributions, working with uncompressed and compressed vector representations, respectively. The former exploits query terms and document terms merging to speedup the search phase while jointly reducing the memory footprint. The latter introduces Product Quantization during the document scoring phase and presents a highly efficient filtering step implemented using bit vectors.
Le Reti Neurali Profonde (DNN) sono l’attuale stato dell’arte nel Machine Learning (ML), ma richiedono enormi requisiti computazionali. In questa tesi, proponiamo tre soluzioni per ridurre tali requisiti nei tasks di Learning to Rank (LtR), classificazione delle immagini e multi-term Dense Retrieval (DR).
LtR è il campo del (ML) utilizzato per ordinare i documenti candidati in un motore di ricerca. Viene proposta una metodologia per addestrare reti neurali efficienti ed efficaci per LtR utilizzando il pruning e la knowledge distillation. Inoltre, vengono sviluppati dei predittori analitici che stimano la latenza di reti neurali sparse e dense, semplificandonde così la progettazione.
Nella classificazione delle immagini, proponiamo Automatic Prune Binarization (APB), un nuovo framework di compressione che arricchisce l'espressività delle reti binarie con pochi pesi full-precision. Inoltre, progettiamo due algoritmi innovativi di moltiplicazione tra matrici per configurazioni a pochi bit, basati sulle efficienti istruzioni bitwise e logiche della CPU.
Nel multi-term DR, vengono proposti due contributi, rispettivamente per rappresentazioni vettoriali compresse e non compresse. Il primo sfrutta la fusione dei termini di query e documenti per velocizzare la fase di ricerca, riducendo anche la memoria necessaria. Il secondo introduce Product Quantization durante la fase di scoring del documento e presenta una fase di filtraggio efficiente implementata utilizzando bit vectors.
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
