Summary of Investigating Idiomaticity in Word Representations, by Wei He et al.
Investigating Idiomaticity in Word Representations
by Wei He, Tiago Kramer Vieira, Marcos Garcia, Carolina Scarton, Marco Idiart, Aline Villavicencio
First submitted to arxiv on: 4 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed paper investigates how well word representation models can capture the idiomatic meanings of multiword expressions, such as noun compounds. The researchers present a dataset of minimal pairs containing human judgments on idiomaticity, paraphrases, and occurrences in naturalistic contexts. They evaluate various representative models using fine-grained metrics of Affinity and Scaled Similarity to determine their sensitivity to perturbations that may change idiomaticity. The results indicate that current models do not accurately represent idiomaticity, despite superficial similarities. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how well computer models can understand the meanings of phrases like “eager beaver” or “break a leg.” Researchers created a dataset with examples of these phrases and asked humans to rate their level of idiomaticity. They then tested various computer models to see if they could capture these idiomatic meanings. The results show that current models are not very good at understanding idiomatic expressions, even when they seem similar to literal combinations of words. |
