Large language models (LLMs) perform well on (at least) some evaluations of both few-shot multilingual adaptation and reasoning. However, evaluating the intersection of these two skills—multilingual few-shot reasoning—is difficult: even relatively low-resource languages can be found in large training corpora, raising the concern that when we intend to evaluate a model’s ability to generalize to a new language, that language may have in fact been present during the model’s training. If such language contamination has occurred, apparent cases of few-shot reasoning could actually be due to memorization. Towards understanding the capability of models to perform multilingual few-shot reasoning, we propose modeLing, a benchmark of Rosetta stone puzzles. This type of puzzle, originating from competitions called Linguistics Olympiads, contain a small number of sentences in a target language not previously known to the solver. Each sentence is translated to the solver’s language such that the provided sentence pairs uniquely specify a single most reasonable underlying set of rules; solving requires applying these rules to translate new expressions (Figure 1). modeLing languages are chosen to be extremely low-resource such that the risk of training data contamination is low, and unlike prior datasets, it consists entirely of problems written specifically for this work, as a further measure against data leakage. Empirically, we find evidence that popular LLMs do not have data leakage on our benchmark.

In this paper, we detail the methods we used to determine the idiomaticity and plausibility of candidate words or phrases into an instructional text as part of the SemEval Task 7: Identifying Plausible Clarifications of Implicit and Underspecified Phrases in Instructional Texts. Given a set of steps in an instructional text, there are certain phrases that most plausibly fill that spot. We explored various possible architectures, including tree-based methods over GloVe embeddings, ensembled BERT and ELECTRA models, and GPT 2-based infilling methods.

An understanding of humor is an essential component of human-facing NLP systems. In this paper, we investigate several methods for detecting humor in short statements as part of Semeval-2021 Shared Task 7. For Task 1a, we apply an ensemble of fine-tuned pre-trained language models; for Tasks 1b, 1c, and 2a, we investigate various tree-based and linear machine learning models. Our final system achieves an F1-score of 0.9571 (ranked 24 / 58) on Task 1a, an RMSE of 0.5580 (ranked 18 / 50) on Task 1b, an F1-score of 0.5024 (ranked 26 / 36) on Task 1c, and an RMSE of 0.7229 (ranked 45 / 48) on Task 2a.