Related papers: Showing Your Work Doesn't Always Work
Research in natural language processing proceeds, in part, by demonstrating that new models achieve superior performance (e.g., accuracy) on held-out test data, compared to previous results. In this paper, we demonstrate that test-set…
Generalization is arguably the most important goal of statistical language modeling research. Publicly available benchmarks and papers published with an open-source code have been critical to advancing the field. However, it is often very…
Statistical hypothesis testing serves as statistical evidence for scientific innovation. However, if the reported results are intentionally biased, hypothesis testing no longer controls the rate of false discovery. In particular, we study…
Experimental studies are a cornerstone of Machine Learning (ML) research. A common and often implicit assumption is that the study's results will generalize beyond the study itself, e.g., to new data. That is, repeating the same study under…
Deep learning models for semantics are generally evaluated using naturalistic corpora. Adversarial methods, in which models are evaluated on new examples with known semantic properties, have begun to reveal that good performance at these…
It has become a common pattern in our field: One group introduces a language task, exemplified by a dataset, which they argue is challenging enough to serve as a benchmark. They also provide a baseline model for it, which then soon is…
Comparative simulation studies are workhorse tools for benchmarking statistical methods. As with other empirical studies, the success of simulation studies hinges on the quality of their design, execution and reporting. If not conducted…
Synthetic data can improve generalization when real data is scarce, but excessive reliance may introduce distributional mismatches that degrade performance. In this paper, we present a learning-theoretic framework to quantify the trade-off…
Algorithms often have tunable parameters that impact performance metrics such as runtime and solution quality. For many algorithms used in practice, no parameter settings admit meaningful worst-case bounds, so the parameters are made…
In this paper, we propose improvements in how estimation bias, e.g., the tendency towards under-estimating the effort, is measured. The proposed approach emphasizes the need to know what the estimates are meant to represent, i.e., the type…
The currently dominating artificial intelligence and machine learning technology, neural networks, builds on inductive statistical learning. Neural networks of today are information processing systems void of understanding and reasoning…
Machine Learning (ML) is increasingly used across many disciplines with impressive reported results. However, recent studies suggest published performance of ML models are often overoptimistic. Validity concerns are underscored by findings…
This position paper discusses the problem of multilingual evaluation. Using simple statistics, such as average language performance, might inject linguistic biases in favor of dominant language families into evaluation methodology. We argue…
Crowdsourcing has been the prevalent paradigm for creating natural language understanding datasets in recent years. A common crowdsourcing practice is to recruit a small number of high-quality workers, and have them massively generate…
As deep neural networks are highly expressive, it is important to find solutions with small generalization gap (the difference between the performance on the training data and unseen data). Focusing on the stochastic nature of training, we…
This paper tackles the challenge of detecting unreliable behavior in regression algorithms, which may arise from intrinsic variability (e.g., aleatoric uncertainty) or modeling errors (e.g., model uncertainty). First, we formally introduce…
Generalization error predictors (GEPs) aim to predict model performance on unseen distributions by deriving dataset-level error estimates from sample-level scores. However, GEPs often utilize disparate mechanisms (e.g., regressors,…
Performative predictions influence the very outcomes they aim to forecast. We study performative predictions that affect a sample (e.g., only existing users of an app) and/or the whole population (e.g., all potential app users). This raises…
People commonly utilize visualizations not only to examine a given dataset, but also to draw generalizable conclusions about the underlying models or phenomena. Prior research has compared human visual inference to that of an optimal…
There has long been debates on how we could interpret neural networks and understand the decisions our models make. Specifically, why deep neural networks tend to be error-prone when dealing with samples that output low softmax scores. We…