Probing LLMs for Joint Encoding of Linguistic Categories
Large Language Models (LLMs) exhibit impressive performance on a range of NLP tasks, due to the general-purpose linguistic knowledge acquired during pretraining. Existing model interpretability research (Tenney et al., 2019) suggests that a linguistic hierarchy emerges in the LLM layers, with lower layers better suited to solving syntactic tasks and higher layers employed for semantic processing. Yet, little is known about how encodings of different linguistic phenomena interact within the models and to what extent processing of linguistically-related categories relies on the same, shared model representations. In this paper, we propose a framework for testing the joint encoding of linguistic categories in LLMs. Focusing on syntax, we find evidence of joint encoding both at the same (related part-of-speech (POS) classes) and different (POS classes and related syntactic dependency relations) levels of linguistic hierarchy. Our cross-lingual experiments show that the same patterns hold across languages in multilingual LLMs.
Code (1)
Tasks
POSSimilar Papers 제목 키워드 기반
The Low-Dimensional Linear Geometry of Contextualized Word Representations
Black-box probing models can reliably extract linguistic features like tense, number, and syntactic role from pretrained word representations. However, the manner in which these features are encoded in representations re…
A Matter of Framing: The Impact of Linguistic Formalism on Probing Results
Deep pre-trained contextualized encoders like BERT (Delvin et al., 2019) demonstrate remarkable performance on a range of downstream tasks. A recent line of research in probing investigates the linguistic knowledge impli…
Cross-neutralising: Probing for joint encoding of linguistic information in multilingual models
Multilingual sentence encoders are widely used to transfer NLP models across languages. The success of this transfer is, however, dependent on the model's ability to encode the patterns of cross-lingual similarity and va…
SentenceXLM-RProbing Internal Representations of Multi-Word Verbs in Large Language Models
This study investigates the internal representations of verb-particle combinations, called multi-word verbs, within transformer-based large language models (LLMs), specifically examining how these models capture lexical …
SentenceProbing for the Usage of Grammatical Number
A central quest of probing is to uncover how pre-trained models encode a linguistic property within their representations. An encoding, however, might be spurious-i.e., the model might not rely on it when making predicti…