{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/i-am-a-strange-dataset-metalinguistic-tests","title":"I am a Strange Dataset: Metalinguistic Tests for Language Models","arxiv_id":"2401.05300","date":"2024-01-10","proceeding":null,"authors":["Tristan Thrush","Jared Moore","Miguel Monares","Christopher Potts","Douwe Kiela"],"abstract":"Statements involving metalinguistic self-reference (\"This paper has six sections.\") are prevalent in many domains. Can current large language models (LLMs) handle such language? In this paper, we present \"I am a Strange Dataset\", a new dataset for addressing this question. There are two subtasks: generation and verification. In generation, models continue statements like \"The penultimate word in this sentence is\" (where a correct continuation is \"is\"). In verification, models judge the truth of statements like \"The penultimate word in this sentence is sentence.\" (false). We also provide minimally different metalinguistic non-self-reference examples to complement the main dataset by probing for whether models can handle metalinguistic language at all. The dataset is hand-crafted by experts and validated by non-expert annotators. We test a variety of open-source LLMs (7B to 70B parameters) as well as closed-source LLMs through APIs. All models perform close to chance across both subtasks and even on the non-self-referential metalinguistic control data, though we find some steady improvement with model scale. GPT 4 is the only model to consistently do significantly better than chance, and it is still only in the 60% range, while our untrained human annotators score well in the 89-93% range. The dataset and evaluation toolkit are available at https://github.com/TristanThrush/i-am-a-strange-dataset.","url_abs":"https://arxiv.org/abs/2401.05300v2","url_pdf":"https://arxiv.org/pdf/2401.05300v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"i-am-a-strange-dataset-metalinguistic-tests","repo_url":"https://github.com/tristanthrush/i-am-a-strange-dataset","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"sentence","task_name":"Sentence"}],"methods":[{"method_slug":"am","method_name":"AM"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt","method_name":"GPT"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2401.05300","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}