arXiv subjects

Bhaskar Mitra

Publications and source records attributed to Bhaskar Mitra.

At least 37 records · Page 2Linked to original sources

arXiv2025

SynDL: A Large-Scale Synthetic Test Collection for Passage Retrieval

Large-scale test collections play a crucial role in Information Retrieval (IR) research. However, according to the Cranfield paradigm and the research into publicly available datasets, the existing information retrieval research studies are commonly developed on small-scale datasets that rely on human assessors for relevance judgments - a time-intensive and expensive process. Recent studies have shown the strong capability of Large Language Models (LLMs) in producing reliable relevance judgments with human accuracy but at a greatly reduced cost. In this paper, to address the missing large-scale ad-hoc document retrieval dataset, we extend the TREC Deep Learning Track (DL) test collection via additional language model synthetic labels to enable researchers to test and evaluate their search systems at a large scale. Specifically, such a test collection includes more than 1,900 test queries from the previous years of tracks. We compare system evaluation with past human labels from past years and find that our synthetically created large-scale test collection can lead to highly correlated system rankings.

cs.IR↗

arXiv2024

JudgeBlender: Ensembling Judgments for Automatic Relevance Assessment

The effective training and evaluation of retrieval systems require a substantial amount of relevance judgments, which are traditionally collected from human assessors -- a process that is both costly and time-consuming. Large Language Models (LLMs) have shown promise in generating relevance labels for search tasks, offering a potential alternative to manual assessments. Current approaches often rely on a single LLM, such as GPT-4, which, despite being effective, are expensive and prone to intra-model biases that can favour systems leveraging similar models. In this work, we introduce JudgeBlender, a framework that employs smaller, open-source models to provide relevance judgments by combining evaluations across multiple LLMs (LLMBlender) or multiple prompts (PromptBlender). By leveraging the LLMJudge benchmark [18], we compare JudgeBlender with state-of-the-art methods and the top performers in the LLMJudge challenge. Our results show that JudgeBlender achieves competitive performance, demonstrating that very large models are often unnecessary for reliable relevance assessments.

cs.IR↗

arXiv2024

Recall, Robustness, and Lexicographic Evaluation

Although originally developed to evaluate sets of items, recall is often used to evaluate rankings of items, including those produced by recommender, retrieval, and other machine learning systems. The application of recall without a formal evaluative motivation has led to criticism of recall as a vague or inappropriate measure. In light of this debate, we reflect on the measurement of recall in rankings from a formal perspective. Our analysis is composed of three tenets: recall, robustness, and lexicographic evaluation. First, we formally define `recall-orientation' as the sensitivity of a metric to a user interested in finding every relevant item. Second, we analyze recall-orientation from the perspective of robustness with respect to possible content consumers and providers, connecting recall to recent conversations about fair ranking. Finally, we extend this conceptual and theoretical treatment of recall by developing a practical preference-based evaluation method based on lexicographic comparison. Through extensive empirical analysis across three recommendation tasks and 17 information retrieval tasks, we establish that our new evaluation method, lexirecall, has convergent validity (i.e., it is correlated with existing recall metrics) and exhibits substantially higher sensitivity in terms of discriminative power and stability in the presence of missing labels. Our conceptual, theoretical, and empirical analysis substantially deepens our understanding of recall and motivates its adoption through connections to robustness and fairness.

cs.IR↗

arXiv2024

Learning to Extract Structured Entities Using Language Models

Recent advances in machine learning have significantly impacted the field of information extraction, with Language Models (LMs) playing a pivotal role in extracting structured information from unstructured text. Prior works typically represent information extraction as triplet-centric and use classical metrics such as precision and recall for evaluation. We reformulate the task to be entity-centric, enabling the use of diverse metrics that can provide more insights from various perspectives. We contribute to the field by introducing Structured Entity Extraction and proposing the Approximate Entity Set OverlaP (AESOP) metric, designed to appropriately assess model performance. Later, we introduce a new Multistage Structured Entity Extraction (MuSEE) model that harnesses the power of LMs for enhanced effectiveness and efficiency by decomposing the extraction task into multiple stages. Quantitative and human side-by-side evaluations confirm that our model outperforms baselines, offering promising directions for future advancements in structured entity extraction. Our source code and datasets are available at https://github.com/microsoft/Structured-Entity-Extraction.

cs.CL↗

arXiv2024

Report on the 1st Workshop on Large Language Model for Evaluation in Information Retrieval (LLM4Eval 2024) at SIGIR 2024

The first edition of the workshop on Large Language Model for Evaluation in Information Retrieval (LLM4Eval 2024) took place in July 2024, co-located with the ACM SIGIR Conference 2024 in the USA (SIGIR 2024). The aim was to bring information retrieval researchers together around the topic of LLMs for evaluation in information retrieval that gathered attention with the advancement of large language models and generative AI. Given the novelty of the topic, the workshop was focused around multi-sided discussions, namely panels and poster sessions of the accepted proceedings papers.

cs.IR↗

arXiv2024

LLMJudge: LLMs for Relevance Judgments

The LLMJudge challenge is organized as part of the LLM4Eval workshop at SIGIR 2024. Test collections are essential for evaluating information retrieval (IR) systems. The evaluation and tuning of a search system is largely based on relevance labels, which indicate whether a document is useful for a specific search and user. However, collecting relevance judgments on a large scale is costly and resource-intensive. Consequently, typical experiments rely on third-party labelers who may not always produce accurate annotations. The LLMJudge challenge aims to explore an alternative approach by using LLMs to generate relevance judgments. Recent studies have shown that LLMs can generate reliable relevance judgments for search systems. However, it remains unclear which LLMs can match the accuracy of human labelers, which prompts are most effective, how fine-tuned open-source LLMs compare to closed-source LLMs like GPT-4, whether there are biases in synthetically generated data, and if data leakage affects the quality of generated labels. This challenge will investigate these questions, and the collected data will be released as a package to support automatic relevance judgment research in information retrieval and search.

cs.IR↗

arXiv2024

On the Impact of High-Order Harmonic Generation in Electrical Distribution Systems

The modern power grid has seen a rise in the integration of non-linear loads, presenting a significant concern for operators. These loads introduce unwanted harmonics, leading to potential issues such as overheating and improper functioning of circuit breakers. In pursuing a more sustainable grid, the adoption of electric vehicles (EVs) and photovoltaic (PV) systems in residential networks has increased. Understanding and examining the effects of high-order harmonic frequencies beyond $1.5$ kHz is crucial to understanding their impact on the operation and planning of electrical distribution systems under varying nonlinear loading conditions. This study investigates a diverse set of critical power electronic loads within a household modeled using PSCAD/EMTdc, analyzing their unique harmonic spectra. This information is utilized to run the time-series harmonic analysis program in OpenDSS on a modified IEEE 34 bus test system model. The impact of high-order harmonics is quantified using metrics that evaluate total harmonic distortion (THD), transformer harmonic-driven eddy current loss component, and propagation of harmonics from the source to the substation transformer.

eess.SY↗

arXiv2024

Data-driven Modeling for Grid Edge IBRs: A Digital Twin Perspective of User-Defined Models

Recent Odessa disturbance events have brought attention to the challenges associated with the interaction between Inverter-Based Resources (IBRs) and the transmission and distribution system. The NERC event diagnosis report has highlighted several issues, emphasizing the need for continuous performance monitoring of these IBRs by system operators. Key areas of concern include the mismatch of control and protection performance of IBRs between the original equipment manufacturer (OEM)-provided models and field measurements. The inability to replicate the realistic response can result in incorrect reliability and resilience studies. In this paper, we developed an approach on how to emulate the behavior of an IBR using measurement data obtained for system operators to utilize in real-time and long-term planning. Two experiments are conducted in the phasor domain and electromagnetic transients (EMT) domain to emulate the behavior for grid forming and grid following inverters under various operating conditions and the effectiveness of the proposed model is demonstrated in terms of accuracy and ease of utilizing user-defined models (UDMs).

eess.SY↗

arXiv2024

Sociotechnical Implications of Generative Artificial Intelligence for Information Access

Robust access to trustworthy information is a critical need for society with implications for knowledge production, public health education, and promoting informed citizenry in democratic societies. Generative AI technologies may enable new ways to access information and improve effectiveness of existing information retrieval systems but we are only starting to understand and grapple with their long-term social implications. In this chapter, we present an overview of some of the systemic consequences and risks of employing generative AI in the context of information access. We also provide recommendations for evaluation and mitigation, and discuss challenges for future research.

cs.IR↗

arXiv2024

Towards Group-aware Search Success

Traditional measures of search success often overlook the varying information needs of different demographic groups. To address this gap, we introduce a novel metric, named Group-aware Search Success (GA-SS). GA-SS redefines search success to ensure that all demographic groups achieve satisfaction from search outcomes. We introduce a comprehensive mathematical framework to calculate GA-SS, incorporating both static and stochastic ranking policies and integrating user browsing models for a more accurate assessment. In addition, we have proposed Group-aware Most Popular Completion (gMPC) ranking model to account for demographic variances in user intent, aligning more closely with the diverse needs of all user groups. We empirically validate our metric and approach with two real-world datasets: one focusing on query auto-completion and the other on movie recommendations, where the results highlight the impact of stochasticity and the complex interplay among various search success metrics. Our findings advocate for a more inclusive approach in measuring search success, as well as inspiring future investigations into the quality of service of search.

cs.IR↗

arXiv2024

Large language models can accurately predict searcher preferences

Relevance labels, which indicate whether a search result is valuable to a searcher, are key to evaluating and optimising search systems. The best way to capture the true preferences of users is to ask them for their careful feedback on which results would be useful, but this approach does not scale to produce a large number of labels. Getting relevance labels at scale is usually done with third-party labellers, who judge on behalf of the user, but there is a risk of low-quality data if the labeller doesn't understand user needs. To improve quality, one standard approach is to study real users through interviews, user studies and direct feedback, find areas where labels are systematically disagreeing with users, then educate labellers about user needs through judging guidelines, training and monitoring. This paper introduces an alternate approach for improving label quality. It takes careful feedback from real users, which by definition is the highest-quality first-party gold data that can be derived, and develops an large language model prompt that agrees with that data. We present ideas and observations from deploying language models for large-scale relevance labelling at Bing, and illustrate with data from TREC. We have found large language models can be effective, with accuracy as good as human labellers and similar capability to pick the hardest queries, best runs, and best groups. Systematic changes to the prompts make a difference in accuracy, but so too do simple paraphrases. To measure agreement with real searchers needs high-quality "gold" labels, but with these we find that models produce better labels than third-party workers, for a fraction of the cost, and these labels let us train notably better rankers.

cs.IR↗

arXiv2024

Synthetic Test Collections for Retrieval Evaluation

Test collections play a vital role in evaluation of information retrieval (IR) systems. Obtaining a diverse set of user queries for test collection construction can be challenging, and acquiring relevance judgments, which indicate the appropriateness of retrieved documents to a query, is often costly and resource-intensive. Generating synthetic datasets using Large Language Models (LLMs) has recently gained significant attention in various applications. In IR, while previous work exploited the capabilities of LLMs to generate synthetic queries or documents to augment training data and improve the performance of ranking models, using LLMs for constructing synthetic test collections is relatively unexplored. Previous studies demonstrate that LLMs have the potential to generate synthetic relevance judgments for use in the evaluation of IR systems. In this paper, we comprehensively investigate whether it is possible to use LLMs to construct fully synthetic test collections by generating not only synthetic judgments but also synthetic queries. In particular, we analyse whether it is possible to construct reliable synthetic test collections and the potential risks of bias such test collections may exhibit towards LLM-based models. Our experiments indicate that using LLMs it is possible to construct synthetic test collections that can reliably be used for retrieval evaluation.

cs.IR↗

arXiv2024

A Framework for Exploring the Consequences of AI-Mediated Enterprise Knowledge Access and Identifying Risks to Workers

Organisations generate vast amounts of information, which has resulted in a long-term research effort into knowledge access systems for enterprise settings. Recent developments in artificial intelligence, in relation to large language models, are poised to have significant impact on knowledge access. This has the potential to shape the workplace and knowledge in new and unanticipated ways. Many risks can arise from the deployment of these types of AI systems, due to interactions between the technical system and organisational power dynamics. This paper presents the Consequence-Mechanism-Risk framework to identify risks to workers from AI-mediated enterprise knowledge access systems. We have drawn on wide-ranging literature detailing risks to workers, and categorised risks as being to worker value, power, and wellbeing. The contribution of our framework is to additionally consider (i) the consequences of these systems that are of moral import: commodification, appropriation, concentration of power, and marginalisation, and (ii) the mechanisms, which represent how these consequences may take effect in the system. The mechanisms are a means of contextualising risk within specific system processes, which is critical for mitigation. This framework is aimed at helping practitioners involved in the design and deployment of AI-mediated knowledge access systems to consider the risks introduced to workers, identify the precise system mechanisms that introduce those risks and begin to approach mitigation. Future work could apply this framework to other technological systems to promote the protection of workers and other groups.

cs.CY↗

arXiv2024

Hy-DAT: A Tool to Address Hydropower Modeling Gaps Using Interdependency, Efficiency Curves, and Unit Dispatch Models

As the power system continues to be flooded with intermittent resources, it becomes more important to accurately assess the role of hydro and its impact on the power grid. While hydropower generation has been studied for decades, dependency of power generation on water availability and constraints in hydro operation are not well represented in power system models used in the planning and operation of large-scale interconnection studies. There are still multiple modeling gaps that need to be addressed; if not, they can lead to inaccurate operation and planning reliability studies, and consequently to unintentional load shedding or even blackouts. As a result, it is very important that hydropower is represented correctly in both steady-state and dynamic power system studies. In this paper, we discuss the development and use of the Hydrological Dispatch and Analysis Tool (Hy-DAT) as an interactive graphical user interface, that uses a novel methodology to address the hydropower modeling gaps like water availability and interdependency using a database and algorithms to generate accurate representative models for power system simulation.

eess.SP↗

arXiv2024

Result Diversification in Search and Recommendation: A Survey

Diversifying return results is an important research topic in retrieval systems in order to satisfy both the various interests of customers and the equal market exposure of providers. There has been growing attention on diversity-aware research during recent years, accompanied by a proliferation of literature on methods to promote diversity in search and recommendation. However, diversity-aware studies in retrieval systems lack a systematic organization and are rather fragmented. In this survey, we are the first to propose a unified taxonomy for classifying the metrics and approaches of diversification in both search and recommendation, which are two of the most extensively researched fields of retrieval systems. We begin the survey with a brief discussion of why diversity is important in retrieval systems, followed by a summary of the various diversity concerns in search and recommendation, highlighting their relationship and differences. For the survey's main body, we present a unified taxonomy of diversification metrics and approaches in retrieval systems, from both the search and recommendation perspectives. In the later part of the survey, we discuss the open research questions of diversity-aware research in search and recommendation in an effort to inspire future innovations and encourage the implementation of diversity in real-world systems.

cs.IR↗

arXiv2024

DiSK: A Diffusion Model for Structured Knowledge

Structured (dictionary-like) data presents challenges for left-to-right language models, as they can struggle with structured entities for a wide variety of reasons such as formatting and sensitivity to the order in which attributes are presented. Tabular generative models suffer from a different set of limitations such as their lack of flexibility. We introduce Diffusion Models of Structured Knowledge (DiSK) - a new architecture and training approach specialized for structured data. DiSK handles text, categorical, and continuous numerical data using a Gaussian mixture model approach, which allows for improved precision when dealing with numbers. It employs diffusion training to model relationships between properties. Experiments demonstrate DiSK's state-of-the-art performance on tabular data modeling, synthesis, and imputation on over 15 datasets across diverse domains. DiSK provides an effective inductive bias for generative modeling and manipulation of structured data. The techniques we propose could open the door to improved knowledge manipulation in future language models.

cs.LG↗

arXiv2024

Three-Stage Adjusted Regression Forecasting (TSARF) for Software Defect Prediction

Software reliability growth models (SRGM) enable failure data collected during testing. Specifically, nonhomogeneous Poisson process (NHPP) SRGM are the most commonly employed models. While software reliability growth models are important, efficient modeling of complex software systems increases the complexity of models. Increased model complexity presents a challenge in identifying robust and computationally efficient algorithms to identify model parameters and reduces the generalizability of the models. Existing studies on traditional software reliability growth models suggest that NHPP models characterize defect data as a smooth continuous curve and fail to capture changes in the defect discovery process. Therefore, the model fits well under ideal conditions, but it is not adaptable and will only fit appropriately shaped data. Neural networks and other machine learning methods have been applied to greater effect [5], however limited due to lack of large samples of defect data especially at earlier stages of testing.

cs.SE↗

arXiv2023

Gaps in Representations of Hydropower Generation in Steady-State and Dynamic Models

In the evolving power system, where new renewable resources continually displace conventional generation, conventional hydropower resources can be an important asset that helps to maintain reliability and flexibility. Varying climatic patterns do affect the operational pattern of hydropower. This would potentially play a vital role in meeting and delivering energy and meeting climate policy needs. Hydropower is one of the oldest forms of renewable energy resources, however, its dependency on water availability and other constraints are not well represented in power system steady state and dynamic models. This leads to multiple gaps in operations planning especially due to high intermittent renewable generation. Operating constraints and lack of high-quality data often become a barrier to hydropower modeling which leads to inconsistencies in reliability and operational planning studies resulting in unintentional blackouts or unforeseen situations. This paper identifies some of the gaps in hydro-based generation representation in steady-state and dynamic models and provides recommendations for their mitigation.

eess.SY↗