paper-with-me

홈 › Papers

Can Transformers Reason in Fragments of Natural Language?

2022-11-10 · Viktor Schlegel, Kamen V. Pavlov, Ian Pratt-Hartmann

State-of-the-art deep-learning-based approaches to Natural Language Processing (NLP) are credited with various capabilities that involve reasoning with natural language texts. In this paper we carry out a large-scale empirical study investigating the detection of formally valid inferences in controlled fragments of natural language for which the satisfiability problem becomes increasingly complex. We find that, while transformer-based language models perform surprisingly well in these scenarios, a deeper analysis re-veals that they appear to overfit to superficial patterns in the data rather than acquiring the logical principles governing the reasoning in these fragments.

📄 PDF Abstract BibTeX arXiv:2211.05417

Code (1)

schlevik/nlr 공식 구현 pytorch

Tasks

valid

Similar Papers 제목 키워드 기반

Are Transformers Able to Reason by Connecting Separated Knowledge in Training Data?

2025-01-27 · Yutong Yin, Zhaoran Wang

Humans exhibit remarkable compositional reasoning by integrating knowledge from various sources. For example, if someone learns ( B = f(A) ) from one source and ( C = g(B) ) from another, they can deduce ( C=g(B)=g(f(A))…

Transformers in the Service of Description Logic-based Contexts

2023-11-15 · Angelos Poulis, Eleni Tsalapati, Manolis Koubarakis

Recent advancements in transformer-based models have initiated research interests in investigating their ability to learn to perform reasoning tasks. However, most of the contexts used for this purpose are in practice ve…

Question Answering

Probing Natural Language Inference Models through Semantic Fragments

2019-09-16 · Kyle Richardson, Hai Hu, Lawrence S. Moss, Ashish Sabharwal

Do state-of-the-art models for language understanding already have, or can they easily learn, abilities such as boolean coordination, quantification, conditionals, comparatives, and monotonicity reasoning (i.e., reasonin…

Natural Language Inference

Intelligent Analyses on Storytelling for Impact Measurement

2020-11-01 · EMNLP (WNUT) 2020 11 · Koen Kicken, Tessa De Maesschalck, Bart Vanrumste, Tom De Keyser 외

This paper explores how Dutch diary fragments, written by family coaches in the social sector, can be analysed automatically using machine learning techniques to quantitatively measure the impact of social coaching. The …

Language ModellingSentiment Analysis

Modeling meaning: computational interpreting and understanding of natural language fragments

2015-05-29 · Michael Kapustin, Pavlo Kapustin

In this introductory article we present the basics of an approach to implementing computational interpreting of natural language aiming to model the meanings of words and phrases. Unlike other approaches, we attempt to d…

Natural Language Understanding