Related papers: Quantifying the Effect of Test Set Contamination o…
In this paper, we consider contamination by code generation test sets, in particular in their use in modern large language models. We discuss three possible sources of such contamination and show findings supporting each of them: (i) direct…
Coreset selection methods have shown promise in reducing the training data size while maintaining model performance for data-efficient machine learning. However, as many datasets suffer from biases that cause models to learn spurious…
The validity of AI safety evaluations depends on models behaving consistently across controlled and deployment settings. Prior work has identified test-time contextual cues, such as hypothetical scenarios, as a source of verbalized…
We identify label errors in the test sets of 10 of the most commonly-used computer vision, natural language, and audio datasets, and subsequently study the potential for these label errors to affect benchmark results. Errors in test sets…
Automatic syllable stress detection is a crucial component in Computer-Assisted Language Learning (CALL) systems for language learners. Current stress detection models are typically trained on clean speech, which may not be robust in…
Large language models can produce toxic or inappropriate text even for benign inputs, creating risks when deployed at scale. Detoxification is therefore important for safety and user trust, particularly when we want to reduce harmful…
The increasing use of Generative Artificial Intelligence (GAI) tools in education highlights the need to understand their influence on individuals' thinking processes and agency. This research explored 20 university students' interaction…
This study examines the impact of AI on human false memories -- recollections of events that did not occur or deviate from actual occurrences. It explores false memory induction through suggestive questioning in Human-AI interactions,…
Human linguistic capacity is often characterized by compositionality and the generalization it enables -- human learners can produce and comprehend novel complex expressions by composing known parts. Several benchmarks exploit…
Improvements in language model capabilities are often attributed to increasing model size or training data, but in some cases smaller models trained on curated data or with different architectural decisions can outperform larger ones…
The rise of generative AI tools like ChatGPT has significantly reshaped education, sparking debates about their impact on learning outcomes and academic integrity. While prior research highlights opportunities and risks, there remains a…
Improving pretraining data quality and size is known to boost downstream performance, but the role of text complexity--how hard a text is to read--remains less explored. We reduce surface-level complexity (shorter sentences, simpler words,…
AI-text detectors achieve high accuracy on in-domain benchmarks, but often struggle to generalize across different generation conditions such as unseen prompts, model families, or domains. While prior work has reported these generalization…
Synthetic data generation is important to training and evaluating neural models for question answering over knowledge graphs. The quality of the data and the partitioning of the datasets into training, validation and test splits impact the…
Programming students have a widespread access to powerful Generative AI tools like ChatGPT. While this can help understand the learning material and assist with exercises, educators are voicing more and more concerns about an overreliance…
Performance of text classification models tends to drop over time due to changes in data, which limits the lifetime of a pretrained model. Therefore an ability to predict a model's ability to persist over time can help design models that…
Context: Machine learning (ML) may enable effective automated test generation. Objective: We characterize emerging research, examining testing practices, researcher goals, ML techniques applied, evaluation, and challenges. Methods: We…
This study investigates whether assessments fostering higher-order thinking skills can reduce plagiarism involving generative AI tools. Participants completed three tasks of varying complexity in four groups: control, e-textbook, Google,…
GenAI systems such as ChatGPT are increasingly discussed in programming education, but the ways in which the research literature conceptualizes and frames their role remain unclear. This chapter applies text mining to publications indexed…
Large language models are increasingly trained on corpora containing both natural language and non-linguistic data like source code. Aside from aiding programming-related tasks, anecdotal evidence suggests that including code in pretraining…