ACL Findings2023
From chocolate bunny to chocolate crocodile: Do Language Models Understand Noun Compounds?
Albert Coil, Vered Shwartz
Abstract
Noun compound interpretation is the task of expressing a noun compound (e.g. chocolate bunny) in a free-text paraphrase that makes the relationship between the constituent nouns explicit (e.g. bunny-shaped chocolate). We propose modifications to the data and evaluation setup of the standard task (Hendrickx et al., 2013), and show that GPT-3 solves it almost perfectly. We then investigate the task of noun compound conceptualization, i.e. paraphrasing a novel or rare noun compound. E.g., chocolate crocodile is a crocodile-shaped chocolate. This task requires creativity, commonsense, and the ability to generalize knowledge about similar concepts. While GPT-3's performance is not perfect, it is better than that of humans-likely thanks to its access to vast amounts of knowledge, and because conceptual processing is effortful for people (Connell and Lynott, 2012). Finally, we estimate the extent to which GPT-3 is reasoning about the world vs. parroting its training data. We find that the outputs from GPT-3 often have significant overlap with a large web corpus, but that the parroting strategy is less beneficial for novel noun compounds.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
4 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Enhanced Noun-Noun Compound Interpretation through Textual Enrichment cites Can Large Language Models Interpret Noun-Noun Compounds? A Linguistically-Motivated Study on Lexicalized and Novel Compounds
- Memorization or Reasoning? Exploring the Idiom Understanding of LLMs cites Language Models are Few-Shot Learners
- Can Large Language Models Interpret Noun-Noun Compounds? A Linguistically-Motivated Study on Lexicalized and Novel Compounds cites Language Models are Few-Shot Learners
- Language Models are Few-Shot Learners cites The Curious Case of Neural Text Degeneration
- Unsupervised Commonsense Question Answering with Self-Talk cites The Curious Case of Neural Text Degeneration
- Revisiting a Pain in the Neck: A Semantic Reasoning Benchmark for Language Models cites The Curious Case of Neural Text Degeneration
Cited by top-tier papers4
- Enhanced Noun-Noun Compound Interpretation through Textual EnrichmentBingyang Ye, Jingxuan Tu, James PustejovskyEMNLP 2025
- Memorization or Reasoning? Exploring the Idiom Understanding of LLMsJisu Kim, Youngwoo Shin, Uiji Hwang, Jihun Choi et al.EMNLP 2025
- Can Large Language Models Interpret Noun-Noun Compounds? A Linguistically-Motivated Study on Lexicalized and Novel CompoundsGiulia Rambelli, Emmanuele Chersoni, Claudia Collacciani, Marianna BolognesiACL 2024
- Revisiting a Pain in the Neck: A Semantic Reasoning Benchmark for Language ModelsYang Liu, Hongming Li, Melissa Xiaohui Qin, Chao Huang et al.ACL 2026
Builds on9
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
- The Curious Case of Neural Text DegenerationAri Holtzman, Jan Buys, Li Du, Maxwell Forbes et al.ICLR 2020 · 4,112 citations
- Quantifying Memorization Across Neural Language ModelsNicholas Carlini, Daphne Ippolito, Matthew Jagielski, Katherine Lee et al.ICLR 2023 · 158 citations
- Unsupervised Commonsense Question Answering with Self-TalkVered Shwartz, Peter West, Ronan Le Bras, Chandra Bhagavatula et al.EMNLP 2020 · 25 citations
- Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled CorpusJesse Dodge, Maarten Sap, Ana Marasovic, William Agnew et al.EMNLP 2021 · 18 citations
Related papers
- Systematicity in GPT-3's Interpretation of Novel English Noun CompoundsSiyan Li, Riley Carlson, Christopher PottsEMNLP Findings 2022 · 12 citations
- Looking inside Noun Compounds: Unsupervised Prepositional and Free Paraphrasing using Language ModelsGirishkumar Ponkiya, V. Rudra Murthy, Pushpak Bhattacharyya, Girish Keshav PalshikarEMNLP Findings 2020 · 11 citations
- Leveraging the Inductive Bias of Large Language Models for Abstract Textual ReasoningChristopher Michael Rytting, David WingateNeurIPS 2021 · 33 citations
- M3UCD: A Multi-task Multimodal Metaphor Understanding Challenge Dataset for LLMsTianlong Zheng, Yating Yang, Rui Dong, Bo Ma et al.AAAI 2026
- Metaphor Understanding Challenge Dataset for LLMsXiaoyu Tong, Rochelle Choenni, Martha Lewis, Ekaterina ShutovaACL 2024
