Probing Multilingual Sentence Representations With X-Probe
Not on the map. Why?
Abstract (arXiv)
This paper extends the task of probing sentence representations for linguistic insight in a multilingual domain. In doing so, we make two contributions: first, we provide datasets for multilingual probing, derived from Wikipedia, in five languages, viz. English, French, German, Spanish and Russian. Second, we evaluate six sentence encoders for each language, each trained by mapping sentence representations to English sentence representations, using sentences in a parallel corpus. We discover that cross-lingually mapped representations are often better at retaining certain linguistic information than representations derived from English encoders trained on natural language inference (NLI) as a downstream task.
Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). First posted on arXiv 12 Jun 2019.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 5; 2022: 5; 2023: 1; 2024: 1; 2025: 1.
13 papers.
Sort:
- 2022Metaphors in Pre-Trained Language Models: Probing and Generalization Across Datasets and Languagescited by 30
- 2024RuBLiMP: Russian Benchmark of Linguistic Minimal Pairscited by 23
- 2021Shaking Syntactic Trees on the Sesame Street: Multilingual Probing with Controllable Perturbationscited by 10
- 2022SocioProbe: What, When, and Where Language Models Learn about Sociodemographicscited by 7
- 2022Evaluation Benchmarks for Spanish Sentence Representationscited by 6
- 2021RuSentEval: Linguistic Source, Encoder Force!cited by 4
- 2022Universal and Independent: Multilingual Probing Framework for Exhaustive Model Interpretation and Evaluationcited by 3
- 2021Discrete Cosine Transform as Universal Sentence Encodercited by 2
- 2023Analyzing Text Representations by Measuring Task Alignmentcited by 2
- 2021A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoderscited by 1
- 2021Do Multi-Lingual Pre-trained Language Models Reveal Consistent Token Attributions in Different Languages?cited by 0
- 2022Analyzing Text Representations under Tight Annotation Budgets: Measuring Structural Alignmentcited by 0
1 more
- 2025Type and Complexity Signals in Multilingual Question Representations7 Oct 2025
- 2024RuBLiMP: Russian Benchmark of Linguistic Minimal Pairs27 Jun 2024
- 2023Analyzing Text Representations by Measuring Task Alignment31 May 2023
- 2022SocioProbe: What, When, and Where Language Models Learn about Sociodemographics8 Nov 2022
- 2022Universal and Independent: Multilingual Probing Framework for Exhaustive Model Interpretation and Evaluation24 Oct 2022
- 2022Analyzing Text Representations under Tight Annotation Budgets: Measuring Structural Alignment11 Oct 2022
- 2022Evaluation Benchmarks for Spanish Sentence Representations15 Apr 2022
- 2022Metaphors in Pre-Trained Language Models: Probing and Generalization Across Datasets and Languages26 Mar 2022
- 2021Do Multi-Lingual Pre-trained Language Models Reveal Consistent Token Attributions in Different Languages?23 Dec 2021
- 2021Shaking Syntactic Trees on the Sesame Street: Multilingual Probing with Controllable Perturbations28 Sep 2021
- 2021Discrete Cosine Transform as Universal Sentence Encoder2 Jun 2021
- 2021A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoders8 Apr 2021
1 more
- 2021RuSentEval: Linguistic Source, Encoder Force!28 Feb 2021
- 2022Metaphors in Pre-Trained Language Models: Probing and Generalization Across Datasets and Languagescited by 30
- 2024RuBLiMP: Russian Benchmark of Linguistic Minimal Pairscited by 23
- 2021Shaking Syntactic Trees on the Sesame Street: Multilingual Probing with Controllable Perturbationscited by 10
- 2022SocioProbe: What, When, and Where Language Models Learn about Sociodemographicscited by 7
- 2022Evaluation Benchmarks for Spanish Sentence Representationscited by 6
- 2021RuSentEval: Linguistic Source, Encoder Force!cited by 4
- 2022Universal and Independent: Multilingual Probing Framework for Exhaustive Model Interpretation and Evaluationcited by 3
- 2021Discrete Cosine Transform as Universal Sentence Encodercited by 2
- 2023Analyzing Text Representations by Measuring Task Alignmentcited by 2
- 2021A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoderscited by 1
- 2021Do Multi-Lingual Pre-trained Language Models Reveal Consistent Token Attributions in Different Languages?cited by 0
- 2022Analyzing Text Representations under Tight Annotation Budgets: Measuring Structural Alignmentcited by 0
12 most cited of 13 All 13 (JSON)
- 2025Type and Complexity Signals in Multilingual Question Representations7 Oct 2025
- 2024RuBLiMP: Russian Benchmark of Linguistic Minimal Pairs27 Jun 2024
- 2023Analyzing Text Representations by Measuring Task Alignment31 May 2023
- 2022SocioProbe: What, When, and Where Language Models Learn about Sociodemographics8 Nov 2022
- 2022Universal and Independent: Multilingual Probing Framework for Exhaustive Model Interpretation and Evaluation24 Oct 2022
- 2022Analyzing Text Representations under Tight Annotation Budgets: Measuring Structural Alignment11 Oct 2022
- 2022Evaluation Benchmarks for Spanish Sentence Representations15 Apr 2022
- 2022Metaphors in Pre-Trained Language Models: Probing and Generalization Across Datasets and Languages26 Mar 2022
- 2021Do Multi-Lingual Pre-trained Language Models Reveal Consistent Token Attributions in Different Languages?23 Dec 2021
- 2021Shaking Syntactic Trees on the Sesame Street: Multilingual Probing with Controllable Perturbations28 Sep 2021
- 2021Discrete Cosine Transform as Universal Sentence Encoder2 Jun 2021
- 2021A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoders8 Apr 2021
12 newest of 13 All 13 (JSON)