Proceedings of the 1st Workshop on Evaluating Vector-Space Representations for NLP 2016
DOI: 10.18653/v1/w16-2509
|View full text |Cite
|
Sign up to set email alerts
|

Capturing Discriminative Attributes in a Distributional Space: Task Proposal

Abstract: If lexical similarity is not enough to reliably assess how word vectors would perform on various specific tasks, we need other ways of evaluating semantic representations. We propose a new task, which consists in extracting semantic differences using distributional models: given two words, what is the difference between their meanings? We present two proof of concept datasets for this task and outline how it may be performed.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
17
0
2

Year Published

2018
2018
2023
2023

Publication Types

Select...
5
2
2

Relationship

1
8

Authors

Journals

citations
Cited by 14 publications
(19 citation statements)
references
References 11 publications
0
17
0
2
Order By: Relevance
“…We therefore select these embeddings for further experiments. Table 2: Averaged F-Score across GloVe Dimensions between our 2-step unsupervised system and the baseline from Krebs and Paperno (2016), for word vectors of size 50, 100, 200 and 300.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…We therefore select these embeddings for further experiments. Table 2: Averaged F-Score across GloVe Dimensions between our 2-step unsupervised system and the baseline from Krebs and Paperno (2016), for word vectors of size 50, 100, 200 and 300.…”
Section: Resultsmentioning
confidence: 99%
“…For example, we would expect the word car to have high semantic similarity with truck and with vehicle in distributional vector spaces, while the relation between car and truck differs from the relation between car and vehicle. In addition, popular datasets for similarity tasks are small, and similarity annotations are subjective with low inter-annotator agreement (Krebs and Paperno, 2016).…”
Section: Introductionmentioning
confidence: 99%
“…When creating the dataset, we started from the approach that Lazaridou et al (2016) used for visual discriminating attribute identification, followed by manual filtering for the test and validation data. Dataset creation consisted of three phases: As an initial source of data, we used the feature norms collected by McRae et al (2005) and created a pilot dataset (Krebs and Paperno, 2016). This set was then reverified and manually extended to improve the quality and the variety of the data.…”
Section: Data Collection and Annotationmentioning
confidence: 99%
“…As emphasized by Krebs and Paperno (2016), this non-trivial task has numerous applications, such as automatized lexicography, conversational agents or machine translation.…”
Section: Introduction and Related Workmentioning
confidence: 99%