Learning from Impairment: Leveraging Insights from Clinical Linguistics in Language Modelling Research

TL;DR

The paper addresses how language models can benefit from insights in clinical linguistics and aphasia rehabilitation to develop human-inspired learning and evaluation frameworks. It surveys three linguistically grounded aphasia therapies—MT, TUF, and SSP—and derives complexity hierarchies and training principles that can guide LM curricula and metrics. The key contributions include proposing evaluation protocols based on incremental syntactic complexity to reveal model weaknesses and inform generalization, as well as learning strategies such as curriculum learning and linguistically informed pre-training, potentially augmented by multimodal inputs. This approach aims to yield cognitively plausible, data-efficient NLP systems with improved generalization and robust handling of complex syntax.

Abstract

This position paper investigates the potential of integrating insights from language impairment research and its clinical treatment to develop human-inspired learning strategies and evaluation frameworks for language models (LMs). We inspect the theoretical underpinnings underlying some influential linguistically motivated training approaches derived from neurolinguistics and, particularly, aphasiology, aimed at enhancing the recovery and generalization of linguistic skills in aphasia treatment, with a primary focus on those targeting the syntactic domain. We highlight how these insights can inform the design of rigorous assessments for LMs, specifically in their handling of complex syntactic phenomena, as well as their implications for developing human-like learning strategies, aligning with efforts to create more sustainable and cognitively plausible natural language processing (NLP) models.