Retrieval-augmented generation (RAG) has become integral to large language models (LLMs), particularly for conversational AI systems where user questions may reference knowledge beyond the LLMs' training cutoff. However, many natural user questions lack well-defined answers, either due to limited domain knowledge or because the retrieval system returns documents that are relevant in appearance but uninformative in content. In such cases, LLMs often produce hallucinated answers without flagging them. While recent work has largely focused on questions with false premises, we study out-of-scope questions, where the retrieved document appears semantically similar to the question but lacks the necessary information to answer it. In this paper, we propose a guided hallucination-based approach ELOQ . https://github.com/zhiyuanpeng/ELOQ.git, for automatically generating a diverse set of out-of-scope questions from post-cutoff documents, followed by human verification to ensure quality. We use this dataset to evaluate several LLMs on their ability to detect out-of-scope questions and generate appropriate responses. Finally, we introduce an improved detection method that enhances the reliability of LLM-based question-answering systems in handling out-of-scope questions.

Deep neural networks have seen great success in recent years; however, training a deep model is often challenging as its performance heavily depends on the hyper-parameters used. In addition, finding the optimal hyper-parameter configuration, even with state-of-the-art (SOTA) hyper-parameter optimization (HPO) algorithms, can be time-consuming, requiring multiple training runs over the entire datasetfor different possible sets of hyper-parameters. Our central insight is that using an informative subset of the dataset for model training runs involved in hyper-parameter optimization, allows us to find the optimal hyper-parameter configuration significantly faster. In this work, we propose AUTOMATA, a gradient-based subset selection framework for hyper-parameter tuning. We empirically evaluate the effectiveness of AUTOMATA in hyper-parameter tuning through several experiments on real-world datasets in the text, vision, and tabular domains. Our experiments show that using gradient-based data subsets for hyper-parameter tuning achieves significantly faster turnaround times and speedups of 3×-30× while achieving comparable performance to the hyper-parameters found using the entire dataset.

Building models for natural language processing (NLP) tasks remains a daunting task for many, requiring significant technical expertise, efforts, and resources. In this demonstration, we present AutoText, an end-to-end AutoAI framework for text, to lower the barrier of entry in building NLP models. AutoText combines state-of-the-art AutoAI optimization techniques and learning algorithms for NLP tasks into a single extensible framework. Through its simple, yet powerful UI, non-AI experts (e.g., domain experts) can quickly generate performant NLP models with support to both control (e.g., via specifying constraints) and understand learned models.

Valuable high-precision data are often published in the form of tables in both scientific and business documents. While humans can easily identify, interpret and contextualize tables, developing general-purpose automated techniques for extraction of information from tables is difficult due to the wide variety of table formats employed across corpora. To extract useful data from tables, data cells must be correctly extracted and linked to all relevant headers, units of measure and in-text references. Table extraction involves identifying the border and cell structure for each document table, while table understanding provides context by linking cells with semantic information inside and outside the table, such as row and column headers, footnotes, titles, and references in surrounding text. The objective of this tutorial is to provide a detailed synopsis of existing approaches for table extraction and understanding, highlight open research problems, and provide an overview of potential applications.

Efficiently computing linear algebra expressions is central to machine learning (ML) systems. Most systems support sparse formats and operations because sparse matrices are ubiquitous and their dense representation can cause prohibitive overheads. Estimating the sparsity of intermediates, however, remains a key challenge when generating execution plans or performing sparse operations. These sparsity estimates are used for cost and memory estimates, format decisions, and result allocation. Existing estimators tend to focus on matrix products only, and struggle to attain good accuracy with low estimation overhead. However, a key observation is that real-world sparse matrices commonly exhibit structural properties such as a single non-zero per row, or columns with varying sparsity. In this paper, we introduce MNC (Matrix Non-zero Count), a remarkably simple, count-based matrix synopsis that exploits these structural properties for efficient, accurate, and general sparsity estimation. We describe estimators and sketch propagation for realistic linear algebra expressions. Our experiments - on a new estimation benchmark called SparsEst - show that the MNC estimator yields good accuracy with very low overhead. This behavior makes MNC practical and broadly applicable in ML systems.

Many machine learning (ML) systems allow the specification of ML algorithms by means of linear algebra programs, and automatically generate efficient execution plans. The opportunities for fused operators---in terms of fused chains of basic operators---are ubiquitous, and include fewer materialized intermediates, fewer scans of inputs, and sparsity exploitation across operators. However, existing fusion heuristics struggle to find good plans for complex operator DAGs or hybrid plans of local and distributed operations. In this paper, we introduce an exact yet practical cost-based optimization framework for fusion plans and describe its end-to-end integration into Apache SystemML. We present techniques for candidate exploration and selection of fusion plans, as well as code generation of local and distributed operations over dense, sparse, and compressed data. Our experiments in SystemML show end-to-end performance improvements of up to 22x, with negligible compilation overhead.

Financial statements report crucial information in tables with complex semantic structure, which are desirable, yet challenging, to interpret automatically. For example, in such tables a row of data cells is often explained by the headers of other rows. In a departure from prior art, we propose a rectangle mining framework for understanding complex tables, which considers rectangular regions rather than individual cells or pairs of cells in a table. We instantiate this framework with ReMine, an algorithm for extracting row header semantics of table, and show that it significantly outperforms prior pair-wise classification approaches on two datasets: (i) a set of manually labeled financial tables from multiple companies, and (ii) the ICDAR 2013 Table Competition dataset.

The rising need for custom machine learning (ML) algorithms and the growing data sizes that require the exploitation of distributed, data-parallel frameworks such as MapReduce or Spark, pose significant productivity challenges to data scientists. Apache SystemML addresses these challenges through declarative ML by (1) increasing the productivity of data scientists as they are able to express custom algorithms in a familiar domain-specific language covering linear algebra primitives and statistical functions, and (2) transparently running these ML algorithms on distributed, data-parallel frameworks by applying cost-based compilation techniques to generate efficient, low-level execution plans with in-memory single-node and large-scale distributed operations. This paper describes SystemML on Apache Spark, end to end, including insights into various optimizer and runtime techniques as well as performance characteristics. We also share lessons learned from porting SystemML to Spark and declarative ML in general. Finally, SystemML is open-source, which allows the database community to leverage it as a testbed for further research.

Meta learning techniques such as cross-validation and ensemble learning are crucial for applying machine learning to real-world use cases. These techniques first generate samples from input data, and then train and evaluate machine learning models on these samples. For meta learning on large datasets, the efficient generation of samples becomes problematic, especially when the data is stored distributed in a block-partitioned representation, and processed on a shared-nothing cluster. We present a novel, parallel algorithm for efficient sample generation from large, block-partitioned datasets in a shared-nothing architecture. This algorithm executes in a single pass over the data, and minimizes inter-machine communication. The algorithm supports a wide variety of sample generation techniques through an embedded user-defined sampling function. We illustrate how to implement distributed sample generation for popular meta learning techniques such as hold-out tests, k-fold cross-validation, and bagging, using our algorithm and present an experimental evaluation on datasets with billions of datapoints.

Analytics on big data range from passenger volume prediction in transportation to customer satisfaction in automotive diagnostic systems, and from correlation analysis in social media data to log analysis in manufacturing. Expressing and running these analytics for varying data characteristics and at scale is challenging. To address these challenges, SystemML implements a declarative, high-level language using an R-like syntax extended with machine-learning-specific constructs, that is compiled to a MapReduce runtime [2]. The language is rich enough to express a wide class of statistical, predictive modeling and machine learning algorithms (Fig. 1). We chose robust algorithms that scale to large, and potentially sparse data with many features.

We present a novel definition of privacy in the framework of offline (retroactive) database query auditing. Given information about the database, a description of sensitive data, and assumptions about users' prior knowledge, our goal is to determine if answering a past user's query could have led to a privacy breach. According to our definition, an audited propertyAis private, given the disclosure of propertyB, if no user can gain confidence inAby learningB, subject to prior knowledge constraints. Privacy is not violated if the disclosure ofBcauses a loss of confidence inA. The new notion of privacy is formalized using the well-known semantics for reasoning about knowledge, where logical properties correspond to sets of possible worlds (databases) that satisfy these properties. Database users are modeled as either possibilistic agents whose knowledge is a set of possible worlds, or as probabilistic agents whose knowledge is a probability distribution on possible worlds. We analyze the new privacy notion, show its relationship with the conventional approach, and derive criteria that allow the auditor to test privacy efficiently in some important cases. In particular, we prove characterization theorems for the possibilistic case, and study in depth the probabilistic case under the assumption that all database records are considered a-priori independent by the user, as well as under more relaxed (or absent) prior-knowledge assumptions. In the probabilistic case we show that for certain families of distributions there is no efficient algorithm to test whether an audited propertyAis private given the disclosure of a propertyB, assumingP≠NP. Nevertheless, for many interesting families, such as the family of product distributions, we obtain algorithms that are efficient both in theory and in practice.

We present a novel definition of privacy in the framework of offline (retroactive) database query auditing. Given information about the database, a description of sensitive data, and assumptions about users' prior knowledge, our goal is to determine if answering a past user's query could have led to a privacy breach. According to our definition, an audited property A is private, given the disclosure of property B, if no user can gain confidence in A by learning B, subject to prior knowledge constraints. Privacy is not violated if the disclosure of B causes a loss of confidence in A. The new notion of privacy is formalized using the well-known semantics for reasoning about knowledge, where logical properties correspond to sets of possible worlds (databases) that satisfy these properties. Database users are modelled as either possibilistic agents whose knowledge is a set of possible worlds, or as probabilistic agents whose knowledge is a probability distribution on possible worlds.We analyze the new privacy notion, show its relationship with the conventional approach, and derive criteria that allow the auditor to test privacy efficiently in some important cases. In particular, we prove characterization theorems for the possibilistic case, and study in depth the probabilistic case under the assumption that all database records are considered a-priori independent by the user, as well as under more relaxed (or absent) prior-knowledge assumptions. In the probabilistic case we show that for certain families of distributions there is no efficient algorithm to test whether an audited property A is private given the disclosure of a property B, assuming P ` NP. Nevertheless, for many interesting families, such as the family of product distributions, we obtain algorithms that are efficient both in theory and in practice.

Numerous widely publicized cases of theft and misuse of private information underscore the need for audit technology to identify the sources of unauthorized disclosure. We present an auditing methodology that ranks potential disclosure sources according to their proximity to the leaked records. Given a sensitive table that contains the disclosed data, our methodology prioritizes by relevance the past queries to the database that could have potentially been used to produce the sensitive table. We provide three conceptually different measures of proximity between the sensitive table and a query result. One measure is inspired by information retrieval in text processing, another is based on statistical record linkage, and the third computes the derivation probability of the sensitive table in a tree-based generative model. We also analyze the characteristics of the three measures and the corresponding ranking algorithms.

There has been increasing interest in the problem of building accurate data mining models over aggregate data, while protecting privacy at the level of individual records. One approach for this problem is to randomize the values in individual records, and only disclose the randomized values. The model is then built over the randomized data, after first compensating for the randomization (at the aggregate level). This approach is potentially vulnerable to privacy breaches: based on the distribution of the data, one may be able to learn with high confidence that some of the randomized records satisfy a specified property, even though privacy is preserved on average.In this paper, we present a new formulation of privacy breaches, together with a methodology, "amplification", for limiting them. Unlike earlier approaches, amplification makes it is possible to guarantee limits on privacy breaches without any knowledge of the distribution of the original data. We instantiate this methodology for the problem of mining association rules, and modify the algorithm from [9] to limit privacy breaches without knowledge of the data distribution. Next, we address the problem that the amount of randomization required to avoid privacy breaches (when mining association rules) results in very long transactions. By using pseudorandom generators and carefully choosing seeds such that the desired items from the original transaction are present in the randomized transaction, we can send just the seed instead of the transaction, resulting in a dramatic drop in communication and storage cost. Finally, we define new information measures that take privacy breaches into account when quantifying the amount of privacy preserved by randomization.

Literature on information integration across databases tacitly assumes that the data in each database can be revealed to the other databases. However, there is an increasing need for sharing information across autonomous entities in such a way that no information apart from the answer to the query is revealed. We formalize the notion of minimal information sharing across private databases, and develop protocols for intersection, equijoin, intersection size, and equijoin size. We also show how new applications can be built using the proposed protocols.

We present a framework for mining association rules from transactions consisting of categorical items where the data has been randomized to preserve privacy of individual transactions. While it is feasible to recover association rules and preserve privacy using a straightforward "uniform" randomization, the discovered rules can unfortunately be exploited to find privacy breaches. We analyze the nature of privacy breaches and propose a class of randomization operators that are much more effective than uniform randomization in limiting the breaches. We derive formulae for an unbiased support estimator and its variance, which allow us to recover itemset supports from randomized datasets, and show how to incorporate these formulae into mining algorithms. Finally, we present experimental results that validate the algorithm by applying it on real datasets.