Qualifying Knowledge and Knowledge Sharing in Multilingual Models

Nicolas Guerin, Emmanuel Chemla

Keywords: Knowledge Retrieval, Pretrained Language Model, LLM Interpretability, Multilingual Models, Cross-linguistic Dataset

Abstract:

Pre-trained language models (PLMs) have demonstrated a remarkable ability to encode factual knowledge. However, the mechanisms underlying how this knowledge is stored and retrieved remain poorly understood, with important implications for AI interpretability and safety. In this paper, we disentangle the multifaceted nature of knowledge: successfully completing a knowledge retrieval task (e.g., “The capital of France is __”) involves mastering underlying concepts (e.g. France, Paris), relationships between these concepts (e.g. capital of), the structure of prompts, including the language of the query. We propose to disentangle these distinct aspects of knowledge and apply this typology to offer a critical view of neuron-level knowledge attribution techniques. For concreteness, we focus on Dai et al.'s (2022) Knowledge Neurons (KNs) across multiple PLMs, testing 10 natural languages and unnatural languages (e.g. Autoprompt).
Our key contributions are twofold: (i) we show that KNs come in different flavors, some indeed encoding entity level concepts, some having a much less transparent, more polysemantic role , and (ii) we uncover an unprecedented overlap in KNs across up to all of the 10 languages we tested, pointing to the existence of a partially unified, language-agnostic retrieval system. To do so, we introduce and release the mParaRel dataset, an extension of ParaRel, featuring prompts and paraphrases for cloze-style knowledge retrieval tasks in parallel over 10 languages.

Primary Area: interpretability and explainable AI

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 11066