Tables and table-based use cases play a crucial role in many important real-world applications, such as spreadsheets, databases, and computational notebooks, which traditionally require expert-level users like data engineers, data analysts, and database administrators to operate. Although LLMs have shown remarkable progress in working with tables (e.g., in spreadsheet and database copilot scenarios), comprehensive benchmarking of such capabilities remains limited. In contrast to an extensive and growing list of NLP benchmarks, evaluations of table-related tasks are scarce, and narrowly focus on tasks like NL-to-SQL and Table-QA, overlooking the broader spectrum of real-world tasks that professional users face. This gap limits our understanding and model progress in this important area.In this work, we introduce MMTU, a large-scale benchmark with over 28K questions across 25 real-world table tasks, designed to comprehensively evaluate models ability to understand, reason, and manipulate real tables at the expert-level. These tasks are drawn from decades’ worth of computer science research on tabular data, with a focus on complex table tasks faced by professional users. We show that MMTU require a combination of skills -- including table understanding, reasoning, and coding -- that remain challenging for today's frontier models, where even frontier reasoning models like OpenAI GPT-5 and DeepSeek R1 score only around 69% and 57% respectively, suggesting significant room for improvement. We highlight key findings in our evaluation using MMTU and hope this benchmark drives further advances in understanding and developing foundation models for structured data processing and analysis.Our code and data are available at https://github.com/MMTU-Benchmark/MMTU and https://huggingface.co/datasets/MMTU-benchmark/MMTU.

We present an approach for estimating the fraction of text in a large corpus which is likely to be substantially modified or produced by a large language model (LLM). Our maximum likelihood model leverages expert-written and AI-generated reference texts to accurately and efficiently examine real-world LLM-use at the corpus level. We apply this approach to a case study of scientific peer review in AI conferences that took place after the release of ChatGPT: *ICLR* 2024, *NeurIPS* 2023, *CoRL* 2023 and *EMNLP* 2023. Our results suggest that between 6.5% and 16.9% of text submitted as peer reviews to these conferences could have been substantially modified by LLMs, i.e. beyond spell-checking or minor writing updates. The circumstances in which generated text occurs offer insight into user behavior: the estimated fraction of LLM-generated text is higher in reviews which report lower confidence, were submitted close to the deadline, and from reviewers who are less likely to respond to author rebuttals. We also observe corpus-level trends in generated text which may be too subtle to detect at the individual level, and discuss the implications of such trends on peer review. We call for future interdisciplinary work to examine how LLM use is changing our information and knowledge practices.

Many recent state-of-the-art results in language tasks were achieved using compound systems that perform multiple Language Model (LM) calls and aggregate their responses. However, there is little understanding of how the number of LM calls -- e.g., when asking the LM to answer each question multiple times and taking a majority vote -- affects such a compound system's performance. In this paper, we initiate the study of scaling properties of compound inference systems. We analyze, theoretically and empirically, how the number of LM calls affects the performance of Vote and Filter-Vote, two of the simplest compound system designs, which aggregate LM responses via majority voting, optionally applying LM filters. We find, surprisingly, that across multiple language tasks, the performance of both Vote and Filter-Vote can first increase but then decrease as a function of the number of LM calls. Our theoretical results suggest that this non-monotonicity is due to the diversity of query difficulties within a task: more LM calls lead to higher performance on "easy" queries, but lower performance on "hard" queries, and non-monotone behavior can emerge when a task contains both types of queries. This insight then allows us to compute, from a small number of samples, the number of LM calls that maximizes system performance, and define an analytical scaling model for both systems. Experiments show that our scaling model can accurately predict the performance of Vote and Filter-Vote systems and thus find the optimal number of LM calls to make.

Machine learning prediction APIs offered by Google, Microsoft, Amazon, and many other providers have been continuously adopted in a plethora of applications, such as visual object detection, natural language comprehension, and speech recognition. Despite the importance of a systematic study and comparison of different APIs over time, this topic is currently under-explored because of the lack of data and user-friendly exploration tools. To address this issue, we present HAPI Explorer (History of API Explorer), an interactive system that offers easy access to millions of instances of commercial API applications collected in three years, prioritize attention on user-defined instance regimes, and explain interesting patterns across different APIs, subpopulations, and time periods via visual and natural languages. HAPI Explorer can facilitate further comprehension and exploitation of ML prediction APIs.

Machine learning research has long focused on models rather than datasets, and prominent datasets are used for common ML tasks without regard to the breadth, difficulty, and faithfulness of the underlying problems. Neglecting the fundamental importance of data has given rise to inaccuracy, bias, and fragility in real-world applications, and research is hindered by saturation across existing dataset benchmarks. In response, we present DataPerf, a community-led benchmark suite for evaluating ML datasets and data-centric algorithms. We aim to foster innovation in data-centric AI through competition, comparability, and reproducibility. We enable the ML community to iterate on datasets, instead of just architectures, and we provide an open, online platform with multiple rounds of challenges to support this iterative development. The first iteration of DataPerf contains five benchmarks covering a wide spectrum of data-centric techniques, tasks, and modalities in vision, speech, acquisition, debugging, and diffusion prompting, and we support hosting new contributed benchmarks from the community. The benchmarks, online evaluation platform, and baseline implementations are open source, and the MLCommons Association will maintain DataPerf to ensure long-term benefits to academia and industry.

Multi-label classification tasks such as OCR and multi-object recognition are a major focus of the growing machine learning as a service industry. While many multi-label APIs are available, it is challenging for users to decide which API to use for their own data and budget, due to the heterogeneity in their prices and performance. Recent work has shown how to efficiently select and combine single label APIs to optimize performance and cost. However, its computation cost is exponential in the number of labels, and is not suitable for settings like OCR. In this work, we propose FrugalMCT, a principled framework that adaptively selects the APIs to use for different data in an online fashion while respecting the user’s budget. It allows combining ML APIs’ predictions for any single data point, and selects the best combination based on an accuracy estimator. We run systematic experiments using ML APIs from Google, Microsoft, Amazon, IBM, Tencent, and other providers for tasks including multi-label image classification, scene text recognition, and named entity recognition. Across these tasks, FrugalMCT can achieve over 90% cost reduction while matching the accuracy of the best single API, or up to 8% better accuracy while matching the best API’s cost.

Commercial ML APIs offered by providers such as Google, Amazon and Microsoft have dramatically simplified ML adoptions in many applications. Numerous companies and academics pay to use ML APIs for tasks such as object detection, OCR and sentiment analysis. Different ML APIs tackling the same task can have very heterogeneous performances. Moreover, the ML models underlying the APIs also evolve over time. As ML APIs rapidly become a valuable marketplace and an integral part of analytics, it is critical to systematically study and compare different APIs with each other and to characterize how individual APIs change over time. However, this practically important topic is currently underexplored due to the lack of data. In this paper, we present HAPI (History of APIs), a longitudinal dataset of 1,761,417 instances of commercial ML API applications (involving APIs from Amazon, Google, IBM, Microsoft and other providers) across diverse tasks including image tagging, speech recognition, and text mining from 2020 to 2022. Each instance consists of a query input for an API (e.g., an image or text) along with the API’s output prediction/annotation and confidence scores. HAPI is the first large-scale dataset of ML API usages and is a unique resource for studying ML as-a-service (MLaaS). As examples of the types of analyses that HAPI enables, we show that ML APIs’ performance changes substantially over time—several APIs’ accuracies dropped on specific benchmark datasets. Even when the API’s aggregate performance stays steady, its error modes can shift across different subtypes of data between 2020 and 2022. Such changes can substantially impact the entire analytics pipelines that use some ML API as a component. We further use HAPI to study commercial APIs’ performance disparities across demographic subgroups over time. HAPI can stimulate more research in the growing field of MLaaS.

Deployed machine learning (ML) models often encounter new user data that differs from their training data. Therefore, estimating how well a given model might perform on the new data is an important step toward reliable ML applications. This is very challenging, however, as the data distribution can change in flexible ways, and we may not have any labels on the new data, which is often the case in monitoring settings. In this paper, we propose a new distribution shift model, Sparse Joint Shift (SJS), which considers the joint shift of both labels and a few features. This unifies and generalizes several existing shift models including label shift and sparse covariate shift, where only marginal feature or label distribution shifts are considered. We describe mathematical conditions under which SJS is identifiable. We further propose SEES, an algorithmic framework to characterize the distribution shift under SJS and to estimate a model’s performance on new data without any labels. We conduct extensive experiments on several real-world datasets with various ML models. Across different datasets and distribution shifts, SEES achieves significant (up to an order of magnitude) shift estimation error improvements over existing approaches.

Offering prediction APIs for fee is a fast growing industry and is an important aspect of machine learning as a service. While many such services are available, the heterogeneity in their price and performance makes it challenging for users to decide which API or combination of APIs to use for their own data and budget. We take a first step towards addressing this challenge by proposing FrugalML, a principled framework that jointly learns the strength and weakness of each API on different data, and performs an efficient optimization to automatically identify the best sequential strategy to adaptively use the available APIs within a budget constraint. Our theoretical analysis shows that natural sparsity in the formulation can be leveraged to make FrugalML efficient. We conduct systematic experiments using ML APIs from Google, Microsoft, Amazon, IBM, Baidu and other providers for tasks including facial emotion recognition, sentiment analysis and speech recognition. Across various tasks, FrugalML can achieve up to 90% cost reduction while matching the accuracy of the best single API, or up to 5% better accuracy while matching the best API’s cost.

Data analytics using machine learning (ML) has become ubiquitous in science, business intelligence, journalism and many other domains. While a lot of work focuses on reducing the training cost, inference runtime and storage cost of ML models, little work studies how to reduce the cost of data acquisition, which potentially leads to a loss of sellers' revenue and buyers' affordability and efficiency. In this paper, we propose a model-based pricing (MBP) framework, which instead of pricing the data, directly prices ML model instances. We first formally describe the desired properties of the MBP framework, with a focus on avoiding arbitrage. Next, we show a concrete realization of the MBP framework via a noise injection approach, which provably satisfies the desired formal properties. Based on the proposed framework, we then provide algorithmic solutions on how the seller can assign prices to models under different market scenarios (such as to maximize revenue). Finally, we conduct extensive experiments, which validate that the MBP framework can provide high revenue to the seller, high affordability to the buyer, and also operate on low runtime cost.

Various domains such as business intelligence and journalism have made many achievements with help of data analytics based on machine learning (ML). While a lot of work has studied how to reduce the cost of training, storing, and deploying ML models, there is little work on eliminating the data collection and purchase cost. Existing data markets provide only simplistic mechanism allowing the sale of fixed datasets with fixed price, which potentially hurts not only ML model availability to buyers with limited budget, but market expansion and thus sellers' revenue as well. In this work, we demonstrate Nimbus, a data market framework for ML model exchange. Instead of pricing data, Nimbus prices ML models directly, which we call model-based pricing (MBP). Through interactive interfaces, the audience can play the role of sellers to vend their own ML models with different price requirements, as well as the role of buyers to purchase ML model instances with different accuracy/budget constraints. We will further demonstrate how much gain of sellers' revenue and buyers' affordability Nimbus can achieve with low runtime cost via both real time and offline results.

Accelerating machine learning (ML) over relational data is a key focus of the database community. While many real-world datasets are multi-table, most ML tools expect single-table inputs, forcing users to materialize joins before ML, leading to data redundancy and runtime waste. Recent works on ''factorized ML'' address such issues by pushing ML through joins. However, they have hitherto been restricted to ML models linear in the feature space, rendering them less effective when users construct non-linear feature interactions such as pairwise products to boost ML accuracy. In this work, we take a first step towards closing this gap by introducing a new abstraction to enable pairwise feature interactions in multi-table data and present an extensive framework of algebraic rewrite rules for factorized LA operators over feature interactions. Our rewrite rules carefully exploit the interplay of the redundancy caused by both joins and interactions. We prototype our framework in Python to build a tool we call MorpheusFI. An extensive empirical evaluation with both synthetic and real datasets shows that MorpheusFI yields up to 5x speedups over materialized execution for a popular second-order gradient method and even an order of magnitude speedups over a popular stochastic gradient method.

Data compression is a popular technique for improving the efficiency of data processing workloads such as SQL queries and more recently, machine learning (ML) with classical batch gradient methods. But the efficacy of such ideas for mini-batch stochastic gradient descent (MGD), arguably the workhorse algorithm of modern ML, is an open question. MGD's unique data access pattern renders prior art, including those designed for batch gradient methods, less effective. We fill this crucial research gap by proposing a new lossless compression scheme we call tuple-oriented compression (TOC) that is inspired by an unlikely source, the string/ text compression scheme Lempel-Ziv-Welch, but tailored to MGD in a way that preserves tuple boundaries within mini-batches. We then present a suite of novel compressed matrix operation execution techniques tailored to the TOC compression scheme that operate directly over the compressed data representation and avoid decompression overheads. An extensive empirical evaluation with real-world datasets shows that TOC consistently achieves substantial compression ratios by up to 51x and reduces runtimes for MGD workloads by up to 10.2x in popular ML systems.

In this paper we study leveraging confidence information induced by adversarial training to reinforce adversarial robustness of a given adversarially trained model. A natural measure of confidence is $\|F(x)\|_\infty$ (i.e. how confident $F$ is about its prediction?). We start by analyzing an adversarial training formulation proposed by Madry et al.. We demonstrate that, under a variety of instantiations, an only somewhat good solution to their objective induces confidence to be a discriminator, which can distinguish between right and wrong model predictions in a neighborhood of a point sampled from the underlying distribution. Based on this, we propose Highly Confident Near Neighbor (HCNN) a framework that combines confidence information and nearest neighbor search, to reinforce adversarial robustness of a base model. We give algorithms in this framework and perform a detailed empirical study. We report encouraging experimental results that support our analysis, and also discuss problems we observed with existing adversarial training.

Distributed model training is vulnerable to byzantine system failures and adversarial compute nodes, i.e., nodes that use malicious updates to corrupt the global model stored at a parameter server (PS). To guarantee some form of robustness, recent work suggests using variants of the geometric median as an aggregation rule, in place of gradient averaging. Unfortunately, median-based rules can incur a prohibitive computational overhead in large-scale settings, and their convergence guarantees often require strong assumptions. In this work, we present DRACO, a scalable framework for robust distributed training that uses ideas from coding theory. In DRACO, each compute node evaluates redundant gradients that are used by the parameter server to eliminate the effects of adversarial updates. DRACO comes with problem-independent robustness guarantees, and the model that it trains is identical to the one trained in the adversary-free setup. We provide extensive experiments on real datasets and distributed setups across a variety of large-scale models, where we show that DRACO is several times, to orders of magnitude faster than median-based approaches.

Distributed implementations of mini-batch stochastic gradient descent (SGD) suffer from communication overheads, attributed to the high frequency of gradient updates inherent in small-batch training. Training with large batches can reduce these overheads; however it besets the convergence of the algorithm and the generalization performance. In this work, we take a first step towards analyzing how the structure (width and depth) of a neural network affects the performance of large-batch training. We present new theoretical results which suggest that--for a fixed number of parameters--wider networks are more amenable to fast large-batch training compared to deeper ones. We provide extensive experiments on residual and fully-connected neural networks which suggest that wider networks can be trained using larger batches without incurring a convergence slow-down, unlike their deeper variants.

Providing machine learning (ML) over relational data is a mainstream requirement for data analytics systems. While almost all ML tools require the input data to be presented as a single table, many datasets are multi-table. This forces data scientists to join those tables first, which often leads to data redundancy and runtime waste. Recent works on "factorized" ML mitigate this issue for a few specific ML algorithms by pushing ML through joins. But their approaches require a manual rewrite of ML implementations. Such piecemeal methods create a massive development overhead when extending such ideas to other ML algorithms. In this paper, we show that it is possible to mitigate this overhead by leveraging a popular formal algebra to represent the computations of many ML algorithms: linear algebra. We introduce a new logical data type to represent normalized data and devise a framework of algebraic rewrite rules to convert a large set of linear algebra operations over denormalized data into operations over normalized data. We show how this enables us to automatically "factorize" several popular ML algorithms, thus unifying and generalizing several prior works. We prototype our framework in the popular ML environment R and an industrial R-over-RDBMS tool. Experiments with both synthetic and real normalized data show that our framework also yields significant speed-ups, up to 36x on real data.

In this paper, a novel public key encryption scheme for large image is presented based on elliptic curve. As a public key encryption scheme, it does not need to exchange and distribute secret keys. Based on elliptic curve discrete logarithm problem (ECDLP), the scheme has high security. The simulation results show that the presented scheme is computationally less complex than ECC (Elliptic Curve Cryptography) and suitable for large image encryption.