Search

While language model (LM)-powered chatbots and generative search engines excel at answering concrete queries, discovering information in the terrain of unknown unknowns remains challenging for users. To emulate the common educational scenario where children/students learn by listening to and participating in conversations of their parents/teachers, we create Collaborative STORM (Co-STORM). Unlike QA systems that require users to ask all the questions, Co-STORM lets users observe and occasionally steer the discourse among several LM agents. The agents ask questions on the user's behalf, allowing the user to discover unknown unknowns serendipitously. To facilitate user interaction, Co-STORM assists users in tracking the discourse by organizing the uncovered information into a dynamic mind map, ultimately generating a comprehensive report as takeaways. For automatic evaluation, we construct the WildSeek dataset by collecting real information-seeking records with user goals. Co-STORM outperforms baseline methods on both discourse trace and report quality. In a further human evaluation, 70% of participants prefer Co-STORM over a search engine, and 78% favor it over a RAG chatbot.

A unique aspect of ColBERT is its use of [MASK] tokens in queries to score documents (query augmentation). Prior work shows [MASK] tokens weighting non-[MASK] query terms, emphasizing certain tokens over others , rather than introducing whole new terms as initially proposed. We begin by demonstrating that a term weighting behavior previously reported for [MASK] tokens in ColBERTv1 holds for ColBERTv2. We then examine the effect of changing the number of [MASK] tokens from zero to up to four times past the query input length used in training, both for first stage retrieval, and for scoring candidates, observing an initial decrease in performance with few [MASK]s, a large increase when enough [MASK]s are added to pad queries to an average length of 32, then a plateau in performance afterwards. Additionally, we compare baseline performance to performance when the query length is extended to 128 tokens, and find that differences are small (e.g., within 1% on various metrics) and generally statistically insignificant, indicating performance does not collapse if ColBERT is presented with more [MASK] tokens than expected., Comment: 5 pages, 3 figures, two tables

Objective: To explore and compare the performance of ChatGPT and other state-of-the-art LLMs on domain-specific NER tasks covering different entity types and domains in TCM against COVID-19 literature. Methods: We established a dataset of 389 articles on TCM against COVID-19, and manually annotated 48 of them with 6 types of entities belonging to 3 domains as the ground truth, against which the NER performance of LLMs can be assessed. We then performed NER tasks for the 6 entity types using ChatGPT (GPT-3.5 and GPT-4) and 4 state-of-the-art BERT-based question-answering (QA) models (RoBERTa, MiniLM, PubMedBERT and SciBERT) without prior training on the specific task. A domain fine-tuned model (GSAP-NER) was also applied for a comprehensive comparison. Results: The overall performance of LLMs varied significantly in exact match and fuzzy match. In the fuzzy match, ChatGPT surpassed BERT-based QA models in 5 out of 6 tasks, while in exact match, BERT-based QA models outperformed ChatGPT in 5 out of 6 tasks but with a smaller F-1 difference. GPT-4 showed a significant advantage over other models in fuzzy match, especially on the entity type of TCM formula and the Chinese patent drug (TFD) and ingredient (IG). Although GPT-4 outperformed BERT-based models on entity type of herb, target, and research method, none of the F-1 scores exceeded 0.5. GSAP-NER, outperformed GPT-4 in terms of F-1 by a slight margin on RM. ChatGPT achieved considerably higher recalls than precisions, particularly in the fuzzy match. Conclusions: The NER performance of LLMs is highly dependent on the entity type, and their performance varies across application scenarios. ChatGPT could be a good choice for scenarios where high recall is favored. However, for knowledge acquisition in rigorous scenarios, neither ChatGPT nor BERT-based QA models are off-the-shelf tools for professional practitioners., Comment: 22 pages with 2 figures

Mathematical information is essential for technical work, but its creation, interpretation, and search are challenging. To help address these challenges, researchers have developed multimodal search engines and mathematical question answering systems. This book begins with a simple framework characterizing the information tasks that people and systems perform as we work to answer math-related questions. The framework is used to organize and relate the other core topics of the book, including interactions between people and systems, representing math formulas in sources, and evaluation. We close with some key questions and concrete directions for future work. This book is intended for use by students, instructors, and researchers, and those who simply wish that it was easier to find and use mathematical information, Comment: [DRAFT] 1st draft

The study of word co-occurrence networks has attracted the attention of researchers due to their potential significance as well as applications. Understanding the structure of word co-occurrence networks is therefore important to fully realize their significance and usages. In past studies, word co-occurrence networks built on well-formed texts have been found to possess certain characteristics, including being small-world, following a two-regime power law distribution, and being generally disassortative. On the flip side, past studies have found that word co-occurrence networks built from ill-formed texts such as microblog posts may behave differently from those built from well-formed documents. While both kinds of word co-occurrence networks are small-world and disassortative, word co-occurrence networks built from ill-formed texts are scale-free and follow the power law distribution instead of the two-regime power law distribution. However, since past studies on the behavior of word co-occurrence networks built from ill-formed texts only investigated English, the universality of such characteristics remains to be seen among different languages. In addition, it is yet to be investigated whether there could be possible similitude/differences between word co-occurrence networks and other potentially comparable networks. This study therefore investigates and compares the structure of word co-occurrence networks and word similarity networks based on Taiwan Mandarin ill-formed internet forum posts and compare them with those built with well-formed judicial judgments, and seeks to find out whether the three aforementioned properties (scale-free, small-world, and disassortative) for ill-formed and well-formed texts are universal among different languages and between word co-occurrence and word similarity networks., Comment: 4 pages, 1 figure, 5 tables

Retrieval Augmented Generation (RAG) is a technique used to augment Large Language Models (LLMs) with contextually relevant, time-critical, or domain-specific information without altering the underlying model parameters. However, constructing RAG systems that can effectively synthesize information from large and diverse set of documents remains a significant challenge. We introduce a novel data-centric RAG workflow for LLMs, transforming the traditional retrieve-then-read system into a more advanced prepare-then-rewrite-then-retrieve-then-read framework, to achieve higher domain expert-level understanding of the knowledge base. Our methodology relies on generating metadata and synthetic Questions and Answers (QA) for each document, as well as introducing the new concept of Meta Knowledge Summary (MK Summary) for metadata-based clusters of documents. The proposed innovations enable personalized user-query augmentation and in-depth information retrieval across the knowledge base. Our research makes two significant contributions: using LLMs as evaluators and employing new comparative performance metrics, we demonstrate that (1) using augmented queries with synthetic question matching significantly outperforms traditional RAG pipelines that rely on document chunking (p < 0.01), and (2) meta knowledge-augmented queries additionally significantly improve retrieval precision and recall, as well as the final answers breadth, depth, relevancy, and specificity. Our methodology is cost-effective, costing less than $20 per 2000 research papers using Claude 3 Haiku, and can be adapted with any fine-tuning of either the language or embedding models to further enhance the performance of end-to-end RAG pipelines., Comment: Accepted in Workshop on Generative AI for Recommender Systems and Personalization, KDD 2024

Similarity search is the problem of finding in a collection of objects those that are similar to a given query object. It is a fundamental problem in modern applications and the objects considered may be as diverse as locations in space, text documents, images, twitter messages, or trajectories of moving objects. In this paper we are motivated by the latter application. Trajectories are recorded movements of mobile objects such as vehicles, animals, public transportation, or parts of the human body. We propose a novel distance function called DistanceAvg to capture the similarity of such movements. To be practical, it is necessary to provide indexing for this distance measure. Fortunately we do not need to start from scratch. A generic and unifying approach is metric space, which organizes the set of objects solely by a distance (similarity) function with certain natural properties. Our function DistanceAvg is a metric. Although metric indexes have been studied for decades and many such structures are available, they do not offer the best performance with trajectories. In this paper we propose a new design, which outperforms the best existing indexes for kNN queries and is equally good for range queries. It is especially suitable for expensive distance functions as they occur in trajectory similarity search. In many applications, kNN queries are more practical than range queries as it may be difficult to determine an appropriate search radius. Our index provides exact result sets for the given distance function., Comment: 54 pages, 26 figures

Clinical machine learning research and AI driven clinical decision support models rely on clinically accurate labels. Manually extracting these labels with the help of clinical specialists is often time-consuming and expensive. This study tests the feasibility of automatic span- and document-level diagnosis extraction from unstructured Dutch echocardiogram reports. We included 115,692 unstructured echocardiogram reports from the UMCU a large university hospital in the Netherlands. A randomly selected subset was manually annotated for the occurrence and severity of eleven commonly described cardiac characteristics. We developed and tested several automatic labelling techniques at both span and document levels, using weighted and macro F1-score, precision, and recall for performance evaluation. We compared the performance of span labelling against document labelling methods, which included both direct document classifiers and indirect document classifiers that rely on span classification results. The SpanCategorizer and MedRoBERTa$.$nl models outperformed all other span and document classifiers, respectively. The weighted F1-score varied between characteristics, ranging from 0.60 to 0.93 in SpanCategorizer and 0.96 to 0.98 in MedRoBERTa$.$nl. Direct document classification was superior to indirect document classification using span classifiers. SetFit achieved competitive document classification performance using only 10% of the training data. Utilizing a reduced label set yielded near-perfect document classification results. We recommend using our published SpanCategorizer and MedRoBERTa$.$nl models for span- and document-level diagnosis extraction from Dutch echocardiography reports. For settings with limited training data, SetFit may be a promising alternative for document classification., Comment: 28 pages, 5 figures

Range filters are probabilistic data structures that answer approximate range emptiness queries. They aid in avoiding processing empty range queries and have use cases in many application domains such as key-value stores and social web analytics. However, current range filter designs do not support dynamically changing and growing datasets. Moreover, several of these designs also exhibit impractically high false positive rates under correlated workloads, which are common in practice. These impediments restrict the applicability of range filters across a wide range of use cases. We introduce Memento filter, the first range filter to offer dynamicity, fast operations, and a robust false positive rate guarantee for any workload. Memento filter partitions the key universe and clusters its keys according to this partitioning. For each cluster, it stores a fingerprint and a list of key suffixes contiguously. The encoding of these lists makes them amenable to existing dynamic filter structures. Due to the well-defined one-to-one mapping from keys to suffixes, Memento filter supports inserts and deletes and can even expand to accommodate a growing dataset. We implement Memento filter on top of a Rank-and-Select Quotient filter and InfiniFilter and demonstrate that it achieves competitive false positive rates and performance with the state-of-the-art while also providing dynamicity. Due to its dynamicity, Memento filter is the first range filter applicable to B-Trees. We showcase this by integrating Memento filter into WiredTiger, a B-Tree-based key-value store. Memento filter doubles WiredTiger's range query throughput when 50\% of the queries are empty while keeping all other cost metrics unharmed., Comment: 15 pages 13 figures 2 tables In Proceedings of SIGMOD International Conference on Management of Data (2025) (SIGMOD 25), June 22-27, Berlin, Germany

Embedding-based neural retrieval (EBR) is an effective search retrieval method in product search for tackling the vocabulary gap between customer search queries and products. The initial launch of our EBR system at Walmart yielded significant gains in relevance and add-to-cart rates [1]. However, despite EBR generally retrieving more relevant products for reranking, we have observed numerous instances of relevance degradation. Enhancing retrieval performance is crucial, as it directly influences product reranking and affects the customer shopping experience. Factors contributing to these degradations include false positives/negatives in the training data and the inability to handle query misspellings. To address these issues, we present several approaches to further strengthen the capabilities of our EBR model in terms of retrieval relevance. We introduce a Relevance Reward Model (RRM) based on human relevance feedback. We utilize RRM to remove noise from the training data and distill it into our EBR model through a multi-objective loss. In addition, we present the techniques to increase the performance of our EBR model, such as typo-aware training, and semi-positive generation. The effectiveness of our EBR is demonstrated through offline relevance evaluation, online AB tests, and successful deployments to live production. [1] Alessandro Magnani, Feng Liu, Suthee Chaidaroon, Sachin Yadav, Praveen Reddy Suram, Ajit Puthenputhussery, Sijie Chen, Min Xie, Anirudh Kashi, Tony Lee, et al. 2022. Semantic retrieval at walmart. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 3495-3503., Comment: 8 pages, 3 figures, CIKM 2024

In embedding-based retrieval, Approximate Nearest Neighbor (ANN) search enables efficient retrieval of similar items from large-scale datasets. While maximizing recall of relevant items is usually the goal of retrieval systems, a low precision may lead to a poor search experience. Unlike lexical retrieval, which inherently limits the size of the retrieved set through keyword matching, dense retrieval via ANN search has no natural cutoff. Moreover, the cosine similarity scores of embedding vectors are often optimized via contrastive or ranking losses, which make them difficult to interpret. Consequently, relying on top-K or cosine-similarity cutoff is often insufficient to filter out irrelevant results effectively. This issue is prominent in product search, where the number of relevant products is often small. This paper introduces a novel relevance filtering component (called "Cosine Adapter") for embedding-based retrieval to address this challenge. Our approach maps raw cosine similarity scores to interpretable scores using a query-dependent mapping function. We then apply a global threshold on the mapped scores to filter out irrelevant results. We are able to significantly increase the precision of the retrieved set, at the expense of a small loss of recall. The effectiveness of our approach is demonstrated through experiments on both public MS MARCO dataset and internal Walmart product search data. Furthermore, online A/B testing on the Walmart site validates the practical value of our approach in real-world e-commerce settings., Comment: 8 pages, 3 figures, CIKM 2024

We study an emerging and intriguing problem of multimodal temporal event forecasting with large language models. Compared to using text or graph modalities, the investigation of utilizing images for temporal event forecasting has not been fully explored, especially in the era of large language models (LLMs). To bridge this gap, we are particularly interested in two key questions of: 1) why images will help in temporal event forecasting, and 2) how to integrate images into the LLM-based forecasting framework. To answer these research questions, we propose to identify two essential functions that images play in the scenario of temporal event forecasting, i.e., highlighting and complementary. Then, we develop a novel framework, named MM-Forecast. It employs an Image Function Identification module to recognize these functions as verbal descriptions using multimodal large language models (MLLMs), and subsequently incorporates these function descriptions into LLM-based forecasting models. To evaluate our approach, we construct a new multimodal dataset, MidEast-TE-mm, by extending an existing event dataset MidEast-TE-mini with images. Empirical studies demonstrate that our MM-Forecast can correctly identify the image functions, and further more, incorporating these verbal function descriptions significantly improves the forecasting performance. The dataset, code, and prompts are available at https://github.com/LuminosityX/MM-Forecast.

Existing generative retrieval (GR) approaches rely on training-based indexing, i.e., fine-tuning a model to memorise the associations between a query and the document identifier (docid) of a relevant document. Training-based indexing has three limitations: high training overhead, under-utilization of the pre-trained knowledge of large language models (LLMs), and challenges in adapting to a dynamic document corpus. To address the above issues, we propose a novel few-shot indexing-based GR framework (Few-Shot GR). It has a novel few-shot indexing process, where we prompt an LLM to generate docids for all documents in a corpus, ultimately creating a docid bank for the entire corpus. During retrieval, we feed a query to the same LLM and constrain it to generate a docid within the docid bank created during indexing, and then map the generated docid back to its corresponding document. Few-Shot GR relies solely on prompting an LLM without requiring any training, making it more efficient. Moreover, we devise few-shot indexing with one-to-many mapping to further enhance Few-Shot GR. Experiments show that Few-Shot GR achieves superior performance to state-of-the-art GR methods that require heavy training.

Verifying fact-checking claims poses a significant challenge, even for humans. Recent approaches have demonstrated that decomposing claims into relevant questions to gather evidence enhances the efficiency of the fact-checking process. In this paper, we provide empirical evidence showing that this question decomposition can be effectively automated. We demonstrate that smaller generative models, fine-tuned for the question generation task using data augmentation from various datasets, outperform large language models by up to 8%. Surprisingly, in some cases, the evidence retrieved using machine-generated questions proves to be significantly more effective for fact-checking than that obtained from human-written questions. We also perform manual evaluation of the decomposed questions to assess the quality of the questions generated., Comment: Accepted in CIKM 2024 as a short paper 4 pages and 1 page references. Fixed typo in author name

We introduce MOSCITO (MOlecular Dynamics Subspace Clustering with Temporal Observance), a subspace clustering for molecular dynamics data. MOSCITO groups those timesteps of a molecular dynamics trajectory together into clusters in which the molecule has similar conformations. In contrast to state-of-the-art methods, MOSCITO takes advantage of sequential relationships found in time series data. Unlike existing work, MOSCITO does not need a two-step procedure with tedious post-processing, but directly models essential properties of the data. Interpreting clusters as Markov states allows us to evaluate the clustering performance based on the resulting Markov state models. In experiments on 60 trajectories and 4 different proteins, we show that the performance of MOSCITO achieves state-of-the-art performance in a novel single-step method. Moreover, by modeling temporal aspects, MOSCITO obtains better segmentation of trajectories, especially for small numbers of clusters., Comment: Accepted as a research paper at BIOKDD 2024

We introduce tulip agent, an architecture for autonomous LLM-based agents with Create, Read, Update, and Delete access to a tool library containing a potentially large number of tools. In contrast to state-of-the-art implementations, tulip agent does not encode the descriptions of all available tools in the system prompt, which counts against the model's context window, or embed the entire prompt for retrieving suitable tools. Instead, the tulip agent can recursively search for suitable tools in its extensible tool library, implemented exemplarily as a vector store. The tulip agent architecture significantly reduces inference costs, allows using even large tool libraries, and enables the agent to adapt and extend its set of tools. We evaluate the architecture with several ablation studies in a mathematics context and demonstrate its generalizability with an application to robotics. A reference implementation and the benchmark are available at github.com/HRI-EU/tulip_agent., Comment: 19 pages, 4 figures

Recommender systems require the simultaneous optimization of multiple objectives to accurately model user interests, necessitating the application of multi-task learning methods. However, existing multi-task learning methods in recommendations overlook the specific characteristics of recommendation scenarios, falling short in achieving proper gradient balance. To address this challenge, we set the target of multi-task learning as attaining the appropriate magnitude balance and the global direction balance, and propose an innovative methodology named GradCraft in response. GradCraft dynamically adjusts gradient magnitudes to align with the maximum gradient norm, mitigating interference from gradient magnitudes for subsequent manipulation. It then employs projections to eliminate gradient conflicts in directions while considering all conflicting tasks simultaneously, theoretically guaranteeing the global resolution of direction conflicts. GradCraft ensures the concurrent achievement of appropriate magnitude balance and global direction balance, aligning with the inherent characteristics of recommendation scenarios. Both offline and online experiments attest to the efficacy of GradCraft in enhancing multi-task performance in recommendations. The source code for GradCraft can be accessed at https://github.com/baiyimeng/GradCraft., Comment: Accepted by KDD'24

Protein retrieval, which targets the deconstruction of the relationship between sequences, structures and functions, empowers the advancing of biology. Basic Local Alignment Search Tool (BLAST), a sequence-similarity-based algorithm, has proved the efficiency of this field. Despite the existing tools for protein retrieval, they prioritize sequence similarity and probably overlook proteins that are dissimilar but share homology or functionality. In order to tackle this problem, we propose a novel protein retrieval framework that mitigates the bias towards sequence similarity. Our framework initiatively harnesses protein language models (PLMs) to embed protein sequences within a high-dimensional feature space, thereby enhancing the representation capacity for subsequent analysis. Subsequently, an accelerated indexed vector database is constructed to facilitate expedited access and retrieval of dense vectors. Extensive experiments demonstrate that our framework can equally retrieve both similar and dissimilar proteins. Moreover, this approach enables the identification of proteins that conventional methods fail to uncover. This framework will effectively assist in protein mining and empower the development of biology., Comment: 16 pages, 12 figures

The increasing consumption of news online in the 21st century coincided with increased publication of disinformation, biased reporting, hate speech and other unwanted Web content. We describe BiasScanner, an application that aims to strengthen democracy by supporting news consumers with scrutinizing news articles they are reading online. BiasScanner contains a server-side pre-trained large language model to identify biased sentences of news articles and a front-end Web browser plug-in. At the time of writing, BiasScanner can identify and classify more than two dozen types of media bias at the sentence level, making it the most fine-grained model and only deployed application (automatic system in use) of its kind. It was implemented in a light-weight and privacy-respecting manner, and in addition to highlighting likely biased sentence it also provides explanations for each classification decision as well as a summary analysis for each news article. While prior research has addressed news bias detection, we are not aware of any work that resulted in a deployed browser plug-in (c.f. also biasscanner.org for a Web demo)., Comment: 10 pages, 3 figures, 1 table

Quantitative information plays a crucial role in understanding and interpreting the content of documents. Many user queries contain quantities and cannot be resolved without understanding their semantics, e.g., ``car that costs less than $10k''. Yet, modern search engines apply the same ranking mechanisms for both words and quantities, overlooking magnitude and unit information. In this paper, we introduce two quantity-aware ranking techniques designed to rank both the quantity and textual content either jointly or independently. These techniques incorporate quantity information in available retrieval systems and can address queries with numerical conditions equal, greater than, and less than. To evaluate the effectiveness of our proposed models, we introduce two novel quantity-aware benchmark datasets in the domains of finance and medicine and compare our method against various lexical and neural models. The code and data are available under https://github.com/satya77/QuantityAwareRankers.

X-ray images are vital in medical diagnostics, but their effectiveness is limited without clinical context. Radiologists often find chest X-rays insufficient for diagnosing underlying diseases, necessitating comprehensive clinical features and data integration. We present a novel technique to enhance the clinical context through augmentation techniques with clinical tabular data, thereby improving its applicability and reliability in AI medical diagnostics. To address this, we introduce a pioneering approach to clinical data augmentation that employs large language models (LLMs) to generate patient contextual synthetic data. This methodology is crucial for training more robust deep learning models in healthcare. It preserves the integrity of real patient data while enriching the dataset with contextually relevant synthetic features, significantly enhancing model performance. DALL-M uses a three-phase feature generation process: (i) clinical context storage, (ii) expert query generation, and (iii) context-aware feature augmentation. DALL-M generates new, clinically relevant features by synthesizing chest X-ray images and reports. Applied to 799 cases using nine features from the MIMIC-IV dataset, it created an augmented set of 91 features. This is the first work to generate contextual values for existing and new features based on patients' X-ray reports, gender, and age and to produce new contextual knowledge during data augmentation. Empirical validation with machine learning models, including Decision Trees, Random Forests, XGBoost, and TabNET, showed significant performance improvements. Incorporating augmented features increased the F1 score by 16.5% and Precision and Recall by approximately 25%. DALL-M addresses a critical gap in clinical data augmentation, offering a robust framework for generating contextually enriched datasets., Comment: we introduce a pioneering approach to clinical data augmentation that employs large language models (LLMs) to generate patient contextual synthetic data. It preserves the integrity of real patient data while enriching the dataset with contextually relevant synthetic features, significantly enhancing model performance

Manual melody detection is a tedious task requiring high expertise level, while automatic detection is often not expressive or powerful enough. Thus, we present MelodyVis, a visual application designed in collaboration with musicology experts to explore melodic patterns in digital sheet music. MelodyVis features five connected views, including a Melody Operator Graph and a Voicing Timeline. The system utilizes eight atomic operators, such as transposition and mirroring, to capture melody repetitions and variations. Users can start their analysis by manually selecting patterns in the sheet view, and then identifying other patterns based on the selected samples through an interactive exploration process. We conducted a user study to investigate the effectiveness and usefulness of our approach and its integrated melodic operators, including usability and mental load questions. We compared the analysis executed by 25 participants with and without the operators. The study results indicate that the participants could identify at least twice as many patterns with activated operators. MelodyVis allows analysts to steer the analysis process and interpret results. Our study also confirms the usefulness of MelodyVis in supporting common analytical tasks in melodic analysis, with participants reporting improved pattern identification and interpretation. Thus, MelodyVis addresses the limitations of fully-automated approaches, enabling music analysts to step into the analysis process and uncover and understand intricate melodic patterns and transformations in sheet music., Comment: 9+2 pages, 9 figures, preprint, originally submitted to IEEE VIS 23, revision

With the rise of social media platforms, especially X.com (formerly Twitter), there is a growing interest in understanding digital social networks and human digital interactions. This paper presents a comprehensive methodology for extracting, processing, and visually analyzing data from X.com, using a combination of Python and R packages, enhanced by our publicly accessible, customizable code. Our approach compiles a dynamic dataset that captures various interactions: replies, retweets, and mentions. To explore deeper insights, the data is subjected to sentiment analysis and keyword coding, indicating shifts in discourse over time. Our method is structured in three primary phases. Initially, R is employed for pulling data and the formation of social network datasets. Following this, the combination of Python and R is utilized for sentiment analysis and keyword coding, aiming to uncover the underlying emotional shifts and language transitions within topics of discussion. The final phase employs R to visualize the dynamic shifts within these social networks. These visualization tools highlight changes in user interactions and patterns of influence. For a practical demonstration, we analyzed conversations on X.com regarding the controversial proposal to halt AI development, focusing specifically on discussions about ChatGPT. By using keyword searches, leading voices in the debate were identified. Our analysis of sentiment and keywords revealed patterns in emotions and language, while visual tools illustrated the development of network connections and their influence. This study emphasizes the vital role of visual tools in understanding online social dynamics in the digital age., Comment: 22 pages, 2 tables, 8 figures. For associated data and methodology scripts, see this GitHub repository: https://github.com/nrjess/twitter-sna. Submitted to arXiv

Given a collection of Boolean spatial feature types, their instances, a neighborhood relation (e.g., proximity), and a hierarchical taxonomy of the feature types, the goal is to find the subsets of feature types or their parents whose spatial interaction is statistically significant. This problem is for taxonomy-reliant applications such as ecology (e.g., finding new symbiotic relationships across the food chain), spatial pathology (e.g., immunotherapy for cancer), retail, etc. The problem is computationally challenging due to the exponential number of candidate co-location patterns generated by the taxonomy. Most approaches for co-location pattern detection overlook the hierarchical relationships among spatial features, and the statistical significance of the detected patterns is not always considered, leading to potential false discoveries. This paper introduces two methods for incorporating taxonomies and assessing the statistical significance of co-location patterns. The baseline approach iteratively checks the significance of co-locations between leaf nodes or their ancestors in the taxonomy. Using the Benjamini-Hochberg procedure, an advanced approach is proposed to control the false discovery rate. This approach effectively reduces the risk of false discoveries while maintaining the power to detect true co-location patterns. Experimental evaluation and case study results show the effectiveness of the approach., Comment: Accepted in The 16th Conference on Spatial Information Theory (COSIT) 2024

Addressing the "vocabulary mismatch" issue in information retrieval is a central challenge for e-commerce search engines, because product pages often miss important keywords that customers search for. Doc2Query[1] is a popular document-expansion technique that predicts search queries for a document and includes the predicted queries with the document for retrieval. However, this approach can be inefficient for e-commerce search, because the predicted query tokens are often already present in the document. In this paper, we propose Doc2Token, a technique that predicts relevant tokens (instead of queries) that are missing from the document and includes these tokens in the document for retrieval. For the task of predicting missing tokens, we introduce a new metric, "novel ROUGE score". Doc2Token is demonstrated to be superior to Doc2Query in terms of novel ROUGE score and diversity of predictions. Doc2Token also exhibits efficiency gains by reducing both training and inference times. We deployed the feature to production and observed significant revenue gain in an online A/B test, and launched the feature to full traffic on Walmart.com. [1] R. Nogueira, W. Yang, J. Lin, K. Cho, Document expansion by query prediction, arXiv preprint arXiv:1904.08375 (2019), Comment: 9 pages, 1 figure, SIGIR 2024 Workshop on eCommerce

Representation learning in sequential recommendation is critical for accurately modeling user interaction patterns and improving recommendation precision. However, existing approaches predominantly emphasize item-to-item transitions, often neglecting the time intervals between interactions, which are closely related to behavior pattern changes. Additionally, broader interaction attributes, such as item frequency, are frequently overlooked. We found that both sequences with more uniform time intervals and items with higher frequency yield better prediction performance. Conversely, non-uniform sequences exacerbate user interest drift and less-frequent items are difficult to model due to sparse sampling, presenting unique challenges inadequately addressed by current methods. In this paper, we propose UniRec, a novel bidirectional enhancement sequential recommendation method. UniRec leverages sequence uniformity and item frequency to enhance performance, particularly improving the representation of non-uniform sequences and less-frequent items. These two branches mutually reinforce each other, driving comprehensive performance optimization in complex sequential recommendation scenarios. Additionally, we present a multidimensional time module to further enhance adaptability. To the best of our knowledge, UniRec is the first method to utilize the characteristics of uniformity and frequency for feature augmentation. Comparing with eleven advanced models across four datasets, we demonstrate that UniRec outperforms SOTA models significantly. The code is available at https://github.com/Linxi000/UniRec., Comment: 15 pages, 8 figures, accepted by CIKM'24, for source code, see https://github.com/Linxi000/UniRec

Click-through-rate (CTR) prediction has an essential impact on improving user experience and revenue in e-commerce search. With the development of deep learning, graph-based methods are well exploited to utilize graph structure extracted from user behaviors and other information to help embedding learning. However, most of the previous graph-based methods mainly focus on recommendation scenarios, and therefore their graph structures highly depend on item's sequential information from user behaviors, ignoring query's sequential signal and query-item correlation. In this paper, we propose a new approach named Light-weight End-to-End Graph Interest Network (EGIN) to effectively mine users' search interests and tackle previous challenges. (i) EGIN utilizes query and item's correlation and sequential information from the search system to build a heterogeneous graph for better CTR prediction in e-commerce search. (ii) EGIN's graph embedding learning shares the same training input and is jointly trained with CTR prediction, making the end-to-end framework effortless to deploy in large-scale search systems. The proposed EGIN is composed of three parts: query-item heterogeneous graph, light-weight graph sampling, and multi-interest network. The query-item heterogeneous graph captures correlation and sequential information of query and item efficiently by the proposed light-weight graph sampling. The multi-interest network is well designed to utilize graph embedding to capture various similarity relationships between query and item to enhance the final CTR prediction. We conduct extensive experiments on both public and industrial datasets to demonstrate the effectiveness of the proposed EGIN. At the same time, the training cost of graph learning is relatively low compared with the main CTR prediction task, ensuring efficiency in practical applications., Comment: 8 pages, 4 figures

In order to ensure autonomous vehicles are safe for on-road deployment, simulation-based testing has become an integral complement to on-road testing. The rise in simulation testing and validation reflects a growing need to verify that AV behavior is consistent with desired outcomes even in edge case scenarios $-$ which may seldom or never appear in on-road testing data. This raises a critical question: to what extent are AV failures in simulation consistent with data collected from real-world testing? As a result of the gap between simulated and real sensor data (sim-to-real gap), failures in simulation can either be spurious (simulation- or simulator-specific issues) or relevant (safety-critical AV system issues). One possible method for validating if simulated time series failures are consistent with real world time series sensor data could involve retrieving instances of the failure scenario from a real-world time series dataset, in order to understand AV performance in these scenarios. Adopting this strategy, we propose a formal definition of what constitutes a match between a real-world labeled time series data item and a simulated scenario written from a fragment of the Scenic probabilistic programming language for simulation generation. With this definition of a match, we develop a querying algorithm that identifies the subset of a labeled time series dataset matching a given scenario. To allow this approach to be used to verify the safety of other cyber-physical systems (CPS), we present a definition and algorithm for matching scalable beyond the autonomous vehicles domain. Experiments demonstrate the precision and scalability of the algorithm for a set of challenging and uncommon time series scenarios identified from the nuScenes autonomous driving dataset. We include a full system implementation of the querying algorithm freely available for use across a wide range of CPS., Comment: 72 pages, 6 figures, 5 algorithms. Published on https://www2.eecs.berkeley.edu/Pubs/TechRpts/2024/EECS-2024-136.html

A mechanism is described that addresses the fundamental trade off between media producers who want to increase reach and consumers who provide attention based on the rate of utility received, and where overreach negatively impacts that rate. An optimal solution can be achieved when the media source considers the impact of overreach in a cost function used in determining the optimal distribution of content to maximize individual consumer utility and participation. The result is a Nash equilibrium between producer and consumer that is also Pareto efficient. Comparison with the literature on Recommender systems highlights the advantages of the mechanism, including identifying an optimal content volume for the consumer and improvements for optimizing with multiple objectives. A practical algorithm for generating the optimal distribution for each consumer is provided., Comment: Main Paper: 20 pages, Appendix with proofs and additional material: 26 pages. This version fixes typos

Rapidly growing numbers of multilingual news consumers pose an increasing challenge to news recommender systems in terms of providing customized recommendations. First, existing neural news recommenders, even when powered by multilingual language models (LMs), suffer substantial performance losses in zero-shot cross-lingual transfer (ZS-XLT). Second, the current paradigm of fine-tuning the backbone LM of a neural recommender on task-specific data is computationally expensive and infeasible in few-shot recommendation and cold-start setups, where data is scarce or completely unavailable. In this work, we propose a news-adapted sentence encoder (NaSE), domain-specialized from a pretrained massively multilingual sentence encoder (SE). To this end, we construct and leverage PolyNews and PolyNewsParallel, two multilingual news-specific corpora. With the news-adapted multilingual SE in place, we test the effectiveness of (i.e., question the need for) supervised fine-tuning for news recommendation, and propose a simple and strong baseline based on (i) frozen NaSE embeddings and (ii) late click-behavior fusion. We show that NaSE achieves state-of-the-art performance in ZS-XLT in true cold-start and few-shot news recommendation.

This study applies social network analysis (SNA) to map and analyze literary networks in St Petersburg from 1999 to 2019, using data from the 'SPbLitGuide' newsletter. By examining co-participation in literary events, we reveal the dynamics and structures of these networks, identifying key communities and influential figures. Our network graph, consisting of 14,066 nodes and 127,068 edges, represents a highly interconnected and cohesive small-world network with robust local clustering and extensive collaboration. Focusing on core participants, we refined the graph and applied community detection methods to identify distinct groups with specific aesthetic preferences and personal connections. These findings provide insights into the structure and dynamics of literary groups in St. Petersburg and provide a foundation for further research in the digital humanities., Comment: 4 pages, 2 figures, 2 tables. This preprint was accepted at the Digital Humanities Conference 2024

Pharmacovigilance (PV) is essential for drug safety, primarily focusing on adverse event monitoring. Traditionally, accessing safety data required database expertise, limiting broader use. This paper introduces a novel application of Large Language Models (LLMs) to democratize database access for non-technical users. Utilizing OpenAI's GPT-4, we developed a chatbot that generates structured query language (SQL) queries from natural language, bridging the gap between domain knowledge and technical requirements. The proposed application aims for more inclusive and efficient data access, enhancing decision making in drug safety. By providing LLMs with plain language summaries of expert knowledge, our approach significantly improves query accuracy over methods relying solely on database schemas. The application of LLMs in this context not only optimizes PV data analysis, ensuring timely and precise drug safety reporting -- a crucial component in adverse drug reaction monitoring -- but also promotes safer pharmacological practices and informed decision making across various data intensive fields., Comment: 17 pages, 5 figures

The use of generative artificial intelligence (GenAI) in academia is a subjective and hotly debated topic. Currently, there are no agreed guidelines towards the usage of GenAI systems in higher education (HE) and, thus, it is still unclear how to make effective use of the technology for teaching and learning practice. This paper provides an overview of the current state of research on GenAI for teaching and learning in HE. To this end, this study conducted a systematic review of relevant studies indexed by Scopus, using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. The search criteria revealed a total of 625 research papers, of which 355 met the final inclusion criteria. The findings from the review showed the current state and the future trends in documents, citations, document sources/authors, keywords, and co-authorship. The research gaps identified suggest that while some authors have looked at understanding the detection of AI-generated text, it may be beneficial to understand how GenAI can be incorporated into supporting the educational curriculum for assessments, teaching, and learning delivery. Furthermore, there is a need for additional interdisciplinary, multidimensional studies in HE through collaboration. This will strengthen the awareness and understanding of students, tutors, and other stakeholders, which will be instrumental in formulating guidelines, frameworks, and policies for GenAI usage., Comment: 20 pages, 10 figures, article published in Education Sciences

In the realm of personalized recommender systems, the challenge of adapting to evolving user preferences and the continuous influx of new users and items is paramount. Conventional models, typically reliant on a static training-test approach, struggle to keep pace with these dynamic demands. Streaming recommendation, particularly through continual graph learning, has emerged as a novel solution. However, existing methods in this area either rely on historical data replay, which is increasingly impractical due to stringent data privacy regulations; or are inability to effectively address the over-stability issue; or depend on model-isolation and expansion strategies. To tackle these difficulties, we present GPT4Rec, a Graph Prompt Tuning method for streaming Recommendation. Given the evolving user-item interaction graph, GPT4Rec first disentangles the graph patterns into multiple views. After isolating specific interaction patterns and relationships in different views, GPT4Rec utilizes lightweight graph prompts to efficiently guide the model across varying interaction patterns within the user-item graph. Firstly, node-level prompts are employed to instruct the model to adapt to changes in the attributes or properties of individual nodes within the graph. Secondly, structure-level prompts guide the model in adapting to broader patterns of connectivity and relationships within the graph. Finally, view-level prompts are innovatively designed to facilitate the aggregation of information from multiple disentangled views. These prompt designs allow GPT4Rec to synthesize a comprehensive understanding of the graph, ensuring that all vital aspects of the user-item interactions are considered and effectively integrated. Experiments on four diverse real-world datasets demonstrate the effectiveness and efficiency of our proposal., Comment: Accepted by SIGIR 2024

The cold start problem, where new users or items have no interaction history, remains a critical challenge in recommender systems (RS). A common solution involves using Knowledge Graphs (KG) to train entity embeddings or Graph Neural Networks (GNNs). Since KGs incorporate auxiliary data and not just user/item interactions, these methods can make relevant recommendations for cold users or items. Graph Reasoning (GR) methods, however, find paths from users to items to recommend using relations in the KG and, in the context of RS, have been used for interpretability. In this study, we propose GRECS: a framework for adapting GR to cold start recommendations. By utilizing explicit paths starting for users rather than relying only on entity embeddings, GRECS can find items corresponding to users' preferences by navigating the graph, even when limited information about users is available. Our experiments show that GRECS mitigates the cold start problem and outperforms competitive baselines across 5 standard datasets while being explainable. This study highlights the potential of GR for developing explainable recommender systems better suited for managing cold users and items.

Rank-Biased Overlap (RBO) is a similarity measure for indefinite rankings: it is top-weighted, and can be computed when only a prefix of the rankings is known or when they have only some items in common. It is widely used for instance to analyze differences between search engines by comparing the rankings of documents they retrieve for the same queries. In these situations, though, it is very frequent to find tied documents that have the same score. Unfortunately, the treatment of ties in RBO remains superficial and incomplete, in the sense that it is not clear how to calculate it from the ranking prefixes only. In addition, the existing way of dealing with ties is very different from the one traditionally followed in the field of Statistics, most notably found in rank correlation coefficients such as Kendall's and Spearman's. In this paper we propose a generalized formulation for RBO to handle ties, thanks to which we complete the original definitions by showing how to perform prefix evaluation. We also use it to fully develop two variants that align with the ones found in the Statistics literature: one when there is a reference ranking to compare to, and one when there is not. Overall, these three variants provide researchers with flexibility when comparing rankings with RBO, by clearly determining what ties mean, and how they should be treated. Finally, using both synthetic and TREC data, we demonstrate the use of these new tie-aware RBO measures. We show that the scores may differ substantially from the original tie-unaware RBO measure, where ties had to be broken at random or by arbitrary criteria such as by document ID. Overall, these results evidence the need for a proper account of ties in rank similarity measures such as RBO., Comment: 10 pages, 5 figures, 4 tables, SIGIR 2024

When adapting Large Language Models for Recommendation (LLMRec), it is crucial to integrate collaborative information. Existing methods achieve this by learning collaborative embeddings in LLMs' latent space from scratch or by mapping from external models. However, they fail to represent the information in a text-like format, which may not align optimally with LLMs. To bridge this gap, we introduce BinLLM, a novel LLMRec method that seamlessly integrates collaborative information through text-like encoding. BinLLM converts collaborative embeddings from external models into binary sequences -- a specific text format that LLMs can understand and operate on directly, facilitating the direct usage of collaborative information in text-like format by LLMs. Additionally, BinLLM provides options to compress the binary sequence using dot-decimal notation to avoid excessively long lengths. Extensive experiments validate that BinLLM introduces collaborative information in a manner better aligned with LLMs, resulting in enhanced performance. We release our code at https://github.com/zyang1580/BinLLM., Comment: Accepted by ACL 2024

It is often noted that single query-item pair relevance training in search does not capture the customer intent. User intent can be better deduced from a series of engagements (Clicks, ATCs, Orders) in a given search session. We propose a novel method for vectorizing session context for capturing and utilizing context in retrieval and rerank. In the runtime, session embedding is an alternative to query embedding, saved and updated after each request in the session, it can be used for retrieval and ranking. We outline session embedding's solution to session-based intent understanding and its architecture, the background to this line of thought in search and recommendation, detail the methodologies implemented, and finally present the results of an implementation of session embedding for query product type classification. We demonstrate improvements over strategies ignoring session context in the runtime for user intent understanding., Comment: 5 pages, 1 Figure, 5 Tables, SIGIR 2024, LLM for Individuals, Groups, and Society

High relevance of retrieved and re-ranked items to the search query is the cornerstone of successful product search, yet measuring relevance of items to queries is one of the most challenging tasks in product information retrieval, and quality of product search is highly influenced by the precision and scale of available relevance-labelled data. In this paper, we present an array of techniques for leveraging Large Language Models (LLMs) for automating the relevance judgment of query-item pairs (QIPs) at scale. Using a unique dataset of multi-million QIPs, annotated by human evaluators, we test and optimize hyper parameters for finetuning billion-parameter LLMs with and without Low Rank Adaption (LoRA), as well as various modes of item attribute concatenation and prompting in LLM finetuning, and consider trade offs in item attribute inclusion for quality of relevance predictions. We demonstrate considerable improvement over baselines of prior generations of LLMs, as well as off-the-shelf models, towards relevance annotations on par with the human relevance evaluators. Our findings have immediate implications for the growing field of relevance judgment automation in product search., Comment: 10 pages, 1 figure, 11 tables - SIGIR 2024, LLM4Eval

Query recommendation systems are ubiquitous in modern search engines, assisting users in producing effective queries to meet their information needs. However, these systems require a large amount of data to produce good recommendations, such as a large collection of documents to index and query logs. In particular, query logs and user data are not available in cold start scenarios. Query logs are expensive to collect and maintain and require complex and time-consuming cascading pipelines for creating, combining, and ranking recommendations. To address these issues, we frame the query recommendation problem as a generative task, proposing a novel approach called Generative Query Recommendation (GQR). GQR uses an LLM as its foundation and does not require to be trained or fine-tuned to tackle the query recommendation problem. We design a prompt that enables the LLM to understand the specific recommendation task, even using a single example. We then improved our system by proposing a version that exploits query logs called Retriever-Augmented GQR (RA-GQR). RA-GQr dynamically composes its prompt by retrieving similar queries from query logs. GQR approaches reuses a pre-existing neural architecture resulting in a simpler and more ready-to-market approach, even in a cold start scenario. Our proposed GQR obtains state-of-the-art performance in terms of NDCG@10 and clarity score against two commercial search engines and the previous state-of-the-art approach on the Robust04 and ClueWeb09B collections, improving on average the NDCG@10 performance up to ~4% on Robust04 and ClueWeb09B w.r.t the previous best competitor. RA-GQR further improve the NDCG@10 obtaining an increase of ~11%, ~6\% on Robust04 and ClueWeb09B w.r.t the best competitor. Furthermore, our system obtained ~59% of user preferences in a blind user study, proving that our method produces the most engaging queries., Comment: Generating Query Recommendations via LLMs

Query Reformulation (QR) is a set of techniques used to transform a user's original search query to a text that better aligns with the user's intent and improves their search experience. Recently, zero-shot QR has been a promising approach due to its ability to exploit knowledge inherent in large language models. Inspired by the success of ensemble prompting strategies which have benefited other tasks, we investigate if they can improve query reformulation. In this context, we propose two ensemble-based prompting techniques, GenQREnsemble and GenQRFusion which leverage paraphrases of a zero-shot instruction to generate multiple sets of keywords to improve retrieval performance ultimately. We further introduce their post-retrieval variants to incorporate relevance feedback from a variety of sources, including an oracle simulating a human user and a "critic" LLM. We demonstrate that an ensemble of query reformulations can improve retrieval effectiveness by up to 18% on nDCG@10 in pre-retrieval settings and 9% on post-retrieval settings on multiple benchmarks, outperforming all previously reported SOTA results. We perform subsequent analyses to investigate the effects of feedback documents, incorporate domain-specific instructions, filter reformulations, and generate fluent reformulations that might be more beneficial to human searchers. Together, the techniques and the results presented in this paper establish a new state of the art in automated query reformulation for retrieval and suggest promising directions for future research., Comment: Extended Work of GenQREnsemble: Zero-Shot LLM Ensemble Prompting for Generative Query Reformulation, Dhole and Agichtein, ECIR 2024. arXiv admin note: text overlap with arXiv:2404.03746

The proposed research aims to develop an innovative semantic query processing system that enables users to obtain comprehensive information about research works produced by Computer Science (CS) researchers at the Australian National University (ANU). The system integrates Large Language Models (LLMs) with the ANU Scholarly Knowledge Graph (ASKG), a structured repository of all research-related artifacts produced at ANU in the CS field. Each artifact and its parts are represented as textual nodes stored in a Knowledge Graph (KG). To address the limitations of traditional scholarly KG construction and utilization methods, which often fail to capture fine-grained details, we propose a novel framework that integrates the Deep Document Model (DDM) for comprehensive document representation and the KG-enhanced Query Processing (KGQP) for optimized complex query handling. DDM enables a fine-grained representation of the hierarchical structure and semantic relationships within academic papers, while KGQP leverages the KG structure to improve query accuracy and efficiency with LLMs. By combining the ASKG with LLMs, our approach enhances knowledge utilization and natural language understanding capabilities. The proposed system employs an automatic LLM-SPARQL fusion to retrieve relevant facts and textual nodes from the ASKG. Initial experiments demonstrate that our framework is superior to baseline methods in terms of accuracy retrieval and query efficiency. We showcase the practical application of our framework in academic research scenarios, highlighting its potential to revolutionize scholarly knowledge management and discovery. This work empowers researchers to acquire and utilize knowledge from documents more effectively and provides a foundation for developing precise and reliable interactions with LLMs., Comment: for the associated repository, see http://w3id.org/kgcp/KGQP

Machine-learned language models have transformed everyday life: they steer us when we study, drive, manage money. They have the potential to transform our civilization. But they hallucinate. Their realities are virtual. This note provides a high-level overview of language models and outlines a low-level model of learning machines. It turns out that, after they become capable of recognizing hallucinations and dreaming safely, as humans tend to be, the language-learning machines proceed to generate broader systems of false beliefs and self-confirming theories, as humans tend to do., Comment: 100 pages, 64 figures; lecture notes, book draft

The IoT and Business Process Management (BPM) communities co-exist in many shared application domains, such as manufacturing and healthcare. The IoT community has a strong focus on hardware, connectivity and data; the BPM community focuses mainly on finding, controlling, and enhancing the structured interactions among the IoT devices in processes. While the field of Process Mining deals with the extraction of process models and process analytics from process event logs, the data produced by IoT sensors often is at a lower granularity than these process-level events. The fundamental questions about extracting and abstracting process-related data from streams of IoT sensor values are: (1) Which sensor values can be clustered together as part of process events?, (2) Which sensor values signify the start and end of such events?, (3) Which sensor values are related but not essential? This work proposes a framework to semi-automatically perform a set of structured steps to convert low-level IoT sensor data into higher-level process events that are suitable for process mining. The framework is meant to provide a generic sequence of abstract steps to guide the event extraction, abstraction, and correlation, with variation points for plugging in specific analysis techniques and algorithms for each step. To assess the completeness of the framework, we present a set of challenges, how they can be tackled through the framework, and an example on how to instantiate the framework in a real-world demonstration from the field of smart manufacturing. Based on this framework, future research can be conducted in a structured manner through refining and improving individual steps.

In today's digital landscape, where cyber attacks have become the norm, the detection of cyber attacks and threats is critically imperative across diverse domains. Our research presents a new empirical framework for cyber threat modeling, adept at parsing and categorizing cyber-related information from news articles, enhancing real-time vigilance for market stakeholders. At the core of this framework is a fine-tuned BERT model, which we call CANAL - Cyber Activity News Alerting Language Model, tailored for cyber categorization using a novel silver labeling approach powered by Random Forest. We benchmark CANAL against larger, costlier LLMs, including GPT-4, LLaMA, and Zephyr, highlighting their zero to few-shot learning in cyber news classification. CANAL demonstrates superior performance by outperforming all other LLM counterparts in both accuracy and cost-effectiveness. Furthermore, we introduce the Cyber Signal Discovery module, a strategic component designed to efficiently detect emerging cyber signals from news articles. Collectively, CANAL and Cyber Signal Discovery module equip our framework to provide a robust and cost-effective solution for businesses that require agile responses to cyber intelligence., Comment: Published in 2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC), Conference Date: 07-09 February 2024

Large language models (LLMs) have demonstrated remarkable capabilities across various domains, although their susceptibility to hallucination poses significant challenges for their deployment in critical areas such as healthcare. To address this issue, retrieving relevant facts from knowledge graphs (KGs) is considered a promising method. Existing KG-augmented approaches tend to be resource-intensive, requiring multiple rounds of retrieval and verification for each factoid, which impedes their application in real-world scenarios. In this study, we propose Self-Refinement-Enhanced Knowledge Graph Retrieval (Re-KGR) to augment the factuality of LLMs' responses with less retrieval efforts in the medical field. Our approach leverages the attribution of next-token predictive probability distributions across different tokens, and various model layers to primarily identify tokens with a high potential for hallucination, reducing verification rounds by refining knowledge triples associated with these tokens. Moreover, we rectify inaccurate content using retrieved knowledge in the post-processing stage, which improves the truthfulness of generated responses. Experimental results on a medical dataset demonstrate that our approach can enhance the factual capability of LLMs across various foundational models as evidenced by the highest scores on truthfulness.

In large e-commerce platforms, search systems are typically composed of a series of modules, including recall, pre-ranking, and ranking phases. The pre-ranking phase, serving as a lightweight module, is crucial for filtering out the bulk of products in advance for the downstream ranking module. Industrial efforts on optimizing the pre-ranking model have predominantly focused on enhancing ranking consistency, model structure, and generalization towards long-tail items. Beyond these optimizations, meeting the system performance requirements presents a significant challenge. Contrasting with existing industry works, we propose a novel method: a Generalizable and RAnk-ConsistEnt Pre-Ranking Model (GRACE), which achieves: 1) Ranking consistency by introducing multiple binary classification tasks that predict whether a product is within the top-k results as estimated by the ranking model, which facilitates the addition of learning objectives on common point-wise ranking models; 2) Generalizability through contrastive learning of representation for all products by pre-training on a subset of ranking product embeddings; 3) Ease of implementation in feature construction and online deployment. Our extensive experiments demonstrate significant improvements in both offline metrics and online A/B test: a 0.75% increase in AUC and a 1.28% increase in CVR.

We present an interactive Web platform that, given a directed graph, allows identifying the most relevant nodes related to a given query node. Besides well-established algorithms such as PageRank and Personalized PageRank, the demo includes Cyclerank, a novel algorithm that addresses some of their limitations by leveraging cyclic paths to compute personalized relevance scores. Our demo design enables two use cases: (a) algorithm comparison, comparing the results obtained with different algorithms, and (b) dataset comparison, for exploring and gaining insights into a dataset and comparing it with others. We provide 50 pre-loaded datasets from Wikipedia, Twitter, and Amazon and seven algorithms. Users can upload new datasets, and new algorithms can be easily added. By showcasing efficient algorithms to compute relevance scores in directed graphs, our tool helps to uncover hidden relationships within the data, which makes of it a valuable addition to the repertoire of graph analysis algorithms., Comment: 4 pages, 1 figure. To appear at 2024 IEEE 40th International Conference on Data Engineering (ICDE)

Information access systems are getting complex, and our understanding of user behavior during information seeking processes is mainly drawn from qualitative methods, such as observational studies or surveys. Leveraging the advances in sensing technologies, our study aims to characterize user behaviors with physiological signals, particularly in relation to cognitive load, affective arousal, and valence. We conduct a controlled lab study with 26 participants, and collect data including Electrodermal Activities, Photoplethysmogram, Electroencephalogram, and Pupillary Responses. This study examines informational search with four stages: the realization of Information Need (IN), Query Formulation (QF), Query Submission (QS), and Relevance Judgment (RJ). We also include different interaction modalities to represent modern systems, e.g., QS by text-typing or verbalizing, and RJ with text or audio information. We analyze the physiological signals across these stages and report outcomes of pairwise non-parametric repeated-measure statistical tests. The results show that participants experience significantly higher cognitive loads at IN with a subtle increase in alertness, while QF requires higher attention. QS involves demanding cognitive loads than QF. Affective responses are more pronounced at RJ than QS or IN, suggesting greater interest and engagement as knowledge gaps are resolved. To the best of our knowledge, this is the first study that explores user behaviors in a search process employing a more nuanced quantitative analysis of physiological signals. Our findings offer valuable insights into user behavior and emotional responses in information seeking processes. We believe our proposed methodology can inform the characterization of more complex processes, such as conversational information seeking.

We study ranked list truncation (RLT) from a novel "retrieve-then-re-rank" perspective, where we optimize re-ranking by truncating the retrieved list (i.e., trim re-ranking candidates). RLT is crucial for re-ranking as it can improve re-ranking efficiency by sending variable-length candidate lists to a re-ranker on a per-query basis. It also has the potential to improve re-ranking effectiveness. Despite its importance, there is limited research into applying RLT methods to this new perspective. To address this research gap, we reproduce existing RLT methods in the context of re-ranking, especially newly emerged large language model (LLM)-based re-ranking. In particular, we examine to what extent established findings on RLT for retrieval are generalizable to the "retrieve-then-re-rank" setup from three perspectives: (i) assessing RLT methods in the context of LLM-based re-ranking with lexical first-stage retrieval, (ii) investigating the impact of different types of first-stage retrievers on RLT methods, and (iii) investigating the impact of different types of re-rankers on RLT methods. We perform experiments on the TREC 2019 and 2020 deep learning tracks, investigating 8 RLT methods for pipelines involving 3 retrievers and 2 re-rankers. We reach new insights into RLT methods in the context of re-ranking., Comment: Accepted for publication as a long paper at SIGIR 2024