Minimizing the inference cost and latency of foundation models has become a crucial area of research. Optimization approaches include theoretically lossless methods and others without accuracy guarantees like quantization. In all of these cases it is crucial to ensure that the model quality has not degraded. However, even at temperature zero, model generations are not necessarily robust even to theoretically lossless model optimizations due to numerical errors. We thus require statistical tools to decide whether a finite-sample accuracy deviation is an evidence of a model's degradation or whether it can be attributed to (harmless) noise in the evaluation. We propose a statistically sound hypothesis testing framework based on McNemar's test allowing to efficiently detect model degradations, while guaranteeing a controlled rate of false positives. The crucial insight is that we have to confront the model scores on each sample, rather than aggregated on the task level. Furthermore, we propose three approaches to aggregate accuracy estimates across multiple benchmarks into a single decision. We provide an implementation on top of the largely adopted open source LM Evaluation Harness and provide a case study illustrating that the method correctly flags degraded models, while not flagging model optimizations that are provably lossless. We find that with our tests even empirical accuracy degradations of 0.3% can be confidently attributed to actual degradations rather than noise.

Although mathematics is often considered culturally neutral, the way mathematical problems are presented can carry implicit cultural context. Existing benchmarks like GSM8K are predominantly rooted in Western norms, including names, currencies, and everyday scenarios. In this work, we create culturally adapted variants of the GSM8K test set for five regions Africa, India, China, Korea, and Japan using prompt-based transformations followed by manual verification. We evaluate six large language models (LLMs), ranging from 8B to 72B parameters, across five prompting strategies to assess their robustness to cultural variation in math problem presentation. Our findings reveal a consistent performance gap: models perform best on the original US-centric dataset and comparatively worse on culturally adapted versions. However, models with reasoning capabilities are more resilient to these shifts, suggesting that deeper reasoning helps bridge cultural presentation gaps in mathematical tasks

Attributing authorship in the era of large language models (LLMs) is increasingly challenging as machine-generated prose rivals human writing. We benchmark two complementary attribution mechanisms , fixed Style Embeddings and an instruction-tuned LLM judge (GPT-4o) on the Human AI Parallel Corpus, an open dataset of 600 balanced instances spanning six domains (academic, news, fiction, blogs, spoken transcripts, and TV/movie scripts). Each instance contains a human prompt with both a gold continuation and an LLM-generated continuation from either GPT-4o or LLaMA-70B-Instruct. The Style Embedding baseline achieves stronger aggregate accuracy on GPT continuations (82 pct vs. 68 pct). The LLM Judge is slightly better than the Style embeddings on LLaMA continuations (85 pct vs. 81 pct) but the results are not statistically significant. Crucially, the LLM judge significantly outperforms in fiction and academic prose, indicating semantic sensitivity, whereas embeddings dominate in spoken and scripted dialogue, reflecting structural strengths. These complementary patterns highlight attribution as a multidimensional problem requiring hybrid strategies. To support reproducibility we provide code on GitHub and derived data on Hugging Face under the MIT license. This open framework provides a reproducible benchmark for attribution quality assessment in AI-generated content, along with a review of related literature influencing this work.

Despite the recent strides in large language models, studies have underscored the existence of social biases within these systems. In this paper, we delve into the validation and comparison of the ethical biases of LLMs concerning globally discussed and potentially sensitive topics, hypothesizing that these biases may arise from language-specific distinctions. Introducing the Multilingual Sensitive Questions & Answers Dataset (MSQAD), we collected news articles from Human Rights Watch covering 17 topics, and generated socially sensitive questions along with corresponding responses in multiple languages. We scrutinized the biases of these responses across languages and topics, employing two statistical hypothesis tests. The results showed that the null hypotheses were rejected in most cases, indicating biases arising from cross-language differences. It demonstrates that ethical biases in responses are widespread across various languages, and notably, these biases were prevalent even among different LLMs. By making the proposed MSQAD openly available, we aim to facilitate future research endeavors focused on examining cross-language biases in LLMs and their variant models.

Large language models (LLMs) can simulate clinical reasoning based on natural language prompts, but their utility in ophthalmology is largely unexplored. This study evaluated GPT-4's ability to interpret structured textual descriptions of retinal fundus photographs and simulate clinical decisions for diabetic retinopathy (DR) and glaucoma screening, including the impact of adding real or synthetic clinical metadata. We conducted a retrospective diagnostic validation study using 300 annotated fundus images. GPT-4 received structured prompts describing each image, with or without patient metadata. The model was tasked with assigning an ICDR severity score, recommending DR referral, and estimating the cup-to-disc ratio for glaucoma referral. Performance was evaluated using accuracy, macro and weighted F1 scores, and Cohen's kappa. McNemar's test and change rate analysis were used to assess the influence of metadata. GPT-4 showed moderate performance for ICDR classification (accuracy 67.5%, macro F1 0.33, weighted F1 0.67, kappa 0.25), driven mainly by correct identification of normal cases. Performance improved in the binary DR referral task (accuracy 82.3%, F1 0.54, kappa 0.44). For glaucoma referral, performance was poor across all settings (accuracy ~78%, F1 <0.04, kappa <0.03). Metadata inclusion did not significantly affect outcomes (McNemar p > 0.05), and predictions remained consistent across conditions. GPT-4 can simulate basic ophthalmic decision-making from structured prompts but lacks precision for complex tasks. While not suitable for clinical use, LLMs may assist in education, documentation, or image annotation workflows in ophthalmology.

Sarcasm detection is challenging for both humans and machines. This work explores how model characteristics impact sarcasm detection in OpenAI's GPT, and Meta's Llama-2 models, given their strong natural language understanding, and popularity. We evaluate fine-tuned and zero-shot models across various sizes, releases, and hyperparameters. Experiments were conducted on the political and balanced (pol-bal) portion of the popular Self-Annotated Reddit Corpus (SARC2.0) sarcasm dataset. Fine-tuned performance improves monotonically with model size within a model family, while hyperparameter tuning also impacts performance. In the fine-tuning scenario, full precision Llama-2-13b achieves state-of-the-art accuracy and $F_1$-score, both measured at 0.83, comparable to average human performance. In the zero-shot setting, one GPT-4 model achieves competitive performance to prior attempts, yielding an accuracy of 0.70 and an $F_1$-score of 0.75. Furthermore, a model's performance may increase or decline with each release, highlighting the need to reassess performance after each release.

While multilingual large language models generally perform adequately, and sometimes even rival English performance on high-resource languages (HRLs), they often significantly underperform on low-resource languages (LRLs). Among several prompting strategies aiming at bridging the gap, multilingual in-context learning (ICL) has been particularly effective when demonstration in target languages is unavailable. However, there lacks a systematic understanding of when and why it works well. In this work, we systematically analyze multilingual ICL, using demonstrations in HRLs to enhance cross-lingual transfer. We show that demonstrations in mixed HRLs consistently outperform English-only ones across the board, particularly for tasks written in LRLs. Surprisingly, our ablation study shows that the presence of irrelevant non-English sentences in the prompt yields measurable gains, suggesting the effectiveness of multilingual exposure itself. Our results highlight the potential of strategically leveraging multilingual resources to bridge the performance gap for underrepresented languages.

The association of social mobility with a college education has been studied since the early 1950's [1]. Although there are some indications that a college education is not as effective as it once was in helping graduates climb the social ladder [2], it is still the most reliable way of doing so. US News & World Report updated its rankings in 2023 to include social mobility [3], and many institutions of higher education are paying more attention to recruitment of first-generation college students and talented students from disadvantaged backgrounds. With the inclusion of such students in the typical college class comes some important considerations. For example, a student from difficult financial circumstances with an academic background to match the profile of any student an elite institution will have more difficulty paying for textbooks, a laptop, a smartphone, and other items that are almost essential to current college life [2]. As of November 2022, one such item that students from advantaged backgrounds will have access to that those from lower income brackets will not is ChatGPT4 [4]. It currently costs $20 per month for a subscription and has been called a "significant leap forward" compared to ChatGPT3.5 [5], which is free [6]. While use of generative AI is prohibited in some college classrooms, this is hard to police, and many students use it regardless of classroom restrictions [7]. When generative AI is allowed, there is a wide array of platforms from which students can choose.

Abstract: Misinformation can seriously impact society, affecting anything from public opinion to institutional confidence and the political horizon of a state. Fake News (FN) proliferation on online websites and Online Social Networks (OSNs) has increased profusely. Various fact-checking websites include news in English and barely provide information about FN in regional languages. Thus the Urdu FN purveyors cannot be discerned using fact-checking portals. FND in regional and resourceconstrained languages lags due to the lack of limited-sized datasets and legitimate lexical resources. The previous datasets for Urdu FND are limited-sized, domain-restricted, publicly unavailable and not manually verified where the news is translated from English into Urdu. In this paper, we curate and contribute the first largest publicly available dataset for Urdu FND, "Ax-to-Grind Urdu", to bridge the identified gaps and limitations of existing Urdu datasets in the literature. It constitutes 10,083 fake and real news on fifteen domains collected from leading and authentic Urdu newspapers and news channel websites in Pakistan and India. The dataset contains news items in Urdu from the year 2017 to the year 2023. The selected models are originally trained on multilingual large corpora. The results of the proposed model are based on performance metrics, F1-score, accuracy, precision, recall and MCC value. F1-score of 0.924, accuracy of 0.956, precision of 0.942, recall of 0.940 and an MCC value of 0.902 demonstrate the effectiveness of the proposed approach for Urdu FND. Comparison analysis with SOTA ML and DL models and existing Urdu benchmark datasets exhibit that the ensemble model outperforms them for Urdu FND.

Explanations on relational data are hard to verify since the explanation structures are more complex (e.g. graphs). To verify interpretable explanations (e.g. explanations of predictions made in images, text, etc.), typically human subjects are used since it does not necessarily require a lot of expertise. However, to verify the quality of a relational explanation requires expertise and is hard to scale-up. GNNExplainer is arguably one of the most popular explanation methods for Graph Neural Networks. In this paper, we develop an approach where we assess the uncertainty in explanations generated by GNNExplainer. Specifically, we ask the explainer to generate explanations for several counterfactual examples. We generate these examples as symmetric approximations of the relational structure in the original data. From these explanations, we learn a factor graph model to quantify uncertainty in an explanation. Our results on several datasets show that our approach can help verify explanations from GNNExplainer by reliably estimating the uncertainty of a relation specified in the explanation.