paper-with-me

홈 › Papers

Assessing the Capacity of Transformer to Abstract Syntactic Representations: A Contrastive Analysis Based on Long-distance Agreement

2022-12-08 · Bingzhi Li, Guillaume Wisniewski, Benoît Crabbé

The long-distance agreement, evidence for syntactic structure, is increasingly used to assess the syntactic generalization of Neural Language Models. Much work has shown that transformers are capable of high accuracy in varied agreement tasks, but the mechanisms by which the models accomplish this behavior are still not well understood. To better understand transformers' internal working, this work contrasts how they handle two superficially similar but theoretically distinct agreement phenomena: subject-verb and object-past participle agreement in French. Using probing and counterfactual analysis methods, our experiments show that i) the agreement task suffers from several confounders which partially question the conclusions drawn so far and ii) transformers handle subject-verb and object-past participle agreements in a way that is consistent with their modeling in theoretical linguistics.

📄 PDF Abstract BibTeX arXiv:2212.04523

Code (1)

bingzhilee/contrastive_analysis 공식 구현 pytorch

Tasks

counterfactualObject

Similar Papers 제목 키워드 기반

Enriching Transformers with Structured Tensor-Product Representations for Abstractive Summarization

2021-06-02 · NAACL 2021 4 · Yichen Jiang, Asli Celikyilmaz, Paul Smolensky, Paul Soulos 외

Abstractive summarization, the task of generating a concise summary of input documents, requires: (1) reasoning over the source document to determine the salient pieces of information scattered across the long document, …

Abstractive Text Summarization

Assessing the Syntactic Capabilities of Transformer-based Multilingual Language Models

2021-05-10 · Findings (ACL) 2021 8 · Laura Pérez-Mayos, Alba Táboas García, Simon Mille, Leo Wanner

Multilingual Transformer-based language models, usually pretrained on more than 100 languages, have been shown to achieve outstanding results in a wide range of cross-lingual transfer tasks. However, it remains unknown w…

Cross-Lingual Transfer

An expressivity analysis of hierarchical modelling in deep transformers via bounded-depth grammars

2026-06-16 · Vinoth Nandakumar, Qiang Qu, Pramod Thebe, Sakshi Khachariya 외 arxiv

Deep neural networks are widely believed to derive their expressive power from their ability to form \textbf{hierarchical representations}, capturing progressively more abstract and compositional features across layers. …

Discourse structure interacts with reference but not syntax in neural language models

2020-10-10 · CONLL 2020 · Forrest Davis, Marten Van Schijndel

Language models (LMs) trained on large quantities of text have been claimed to acquire abstract linguistic representations. Our work tests the robustness of these abstractions by focusing on the ability of LMs to learn i…

coreference-resolutionCoreference ResolutionLanguage ModelingLanguage Modelling

Mitigating Frequency Bias and Anisotropy in Language Model Pre-Training with Syntactic Smoothing

2024-10-15 · Richard Diehl Martinez, Zebulon Goriely, Andrew Caines, Paula Buttery 외

Language models strongly rely on frequency information because they maximize the likelihood of tokens during pre-training. As a consequence, language models tend to not generalize well to tokens that are seldom seen duri…

Language ModelingLanguage ModellingSentence