One Mind, Many Tongues: A Deep Dive into Language-Agnostic Knowledge Neurons in Large Language Models

• URL: http://arxiv.org/abs/2411.17401v1
• Date: Tue, 26 Nov 2024 13:03:49 GMT
• Title: One Mind, Many Tongues: A Deep Dive into Language-Agnostic Knowledge Neurons in Large Language Models
• Authors: Pengfei Cao, Yuheng Chen, Zhuoran Jin, Yubo Chen, Kang Liu, Jun Zhao,
• Abstract summary: Large language models (LLMs) have learned vast amounts of factual knowledge through self-supervised pre-training on large-scale corpora. LLMs have also demonstrated excellent multilingual capabilities, which can express the learned knowledge in multiple languages.
• Score: 19.58983929459173
• License:
• Abstract: Large language models (LLMs) have learned vast amounts of factual knowledge through self-supervised pre-training on large-scale corpora. Meanwhile, LLMs have also demonstrated excellent multilingual capabilities, which can express the learned knowledge in multiple languages. However, the knowledge storage mechanism in LLMs still remains mysterious. Some researchers attempt to demystify the factual knowledge in LLMs from the perspective of knowledge neurons, and subsequently discover language-agnostic knowledge neurons that store factual knowledge in a form that transcends language barriers. However, the preliminary finding suffers from two limitations: 1) High Uncertainty in Localization Results. Existing study only uses a prompt-based probe to localize knowledge neurons for each fact, while LLMs cannot provide consistent answers for semantically equivalent queries. Thus, it leads to inaccurate localization results with high uncertainty. 2) Lack of Analysis in More Languages. The study only analyzes language-agnostic knowledge neurons on English and Chinese data, without exploring more language families and languages. Naturally, it limits the generalizability of the findings. To address aforementioned problems, we first construct a new benchmark called Rephrased Multilingual LAMA (RML-LAMA), which contains high-quality cloze-style multilingual parallel queries for each fact. Then, we propose a novel method named Multilingual Integrated Gradients with Uncertainty Estimation (MATRICE), which quantifies the uncertainty across queries and languages during knowledge localization. Extensive experiments show that our method can accurately localize language-agnostic knowledge neurons. We also further investigate the role of language-agnostic knowledge neurons in cross-lingual knowledge editing, knowledge enhancement and new knowledge injection.

Related papers

• The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units [16.317199232071232]
  Large language models (LLMs) exhibit remarkable capabilities on not just language tasks, but also various tasks that are not linguistic in nature. In the human brain, neuroscience has identified a core language system that selectively and causally supports language processing. We identify language-selective units within 18 popular LLMs, using the same localization approach that is used in neuroscience.
  arXiv  Detail & Related papers  (2024-11-04T17:09:10Z)
• How Do Multilingual Models Remember? Investigating Multilingual Factual Recall Mechanisms [50.13632788453612]
  Large Language Models (LLMs) store and retrieve vast amounts of factual knowledge acquired during pre-training. The question of how these processes generalize to other languages and multilingual LLMs remains unexplored. We examine when language plays a role in the recall process, uncovering evidence of language-independent and language-dependent mechanisms.
  arXiv  Detail & Related papers  (2024-10-18T11:39:34Z)
• Multilingual Knowledge Editing with Language-Agnostic Factual Neurons [98.73585104789217]
  We investigate how large language models (LLMs) represent multilingual factual knowledge. We find that the same factual knowledge in different languages generally activates a shared set of neurons, which we call language-agnostic factual neurons. Inspired by this finding, we propose a new MKE method by locating and modifying Language-Agnostic Factual Neurons (LAFN) to simultaneously edit multilingual knowledge.
  arXiv  Detail & Related papers  (2024-06-24T08:06:56Z)
• MLaKE: Multilingual Knowledge Editing Benchmark for Large Language Models [65.10456412127405]
  MLaKE is a benchmark for the adaptability of knowledge editing methods across five languages. MLaKE aggregates fact chains from Wikipedia across languages and generates questions in both free-form and multiple-choice. We evaluate the multilingual knowledge editing generalization capabilities of existing methods on MLaKE.
  arXiv  Detail & Related papers  (2024-04-07T15:23:28Z)
• Revealing the Parallel Multilingual Learning within Large Language Models [50.098518799536144]
  In this study, we reveal an in-context learning capability of multilingual large language models (LLMs) By translating the input to several languages, we provide Parallel Input in Multiple Languages (PiM) to LLMs, which significantly enhances their comprehension abilities.
  arXiv  Detail & Related papers  (2024-03-14T03:33:46Z)
• Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models [117.20416338476856]
  Large language models (LLMs) demonstrate remarkable multilingual capabilities without being pre-trained on specially curated multilingual parallel corpora. We propose a novel detection method, language activation probability entropy (LAPE), to identify language-specific neurons within LLMs. Our findings indicate that LLMs' proficiency in processing a particular language is predominantly due to a small subset of neurons.
  arXiv  Detail & Related papers  (2024-02-26T09:36:05Z)
• Unveiling A Core Linguistic Region in Large Language Models [49.860260050718516]
  This paper conducts an analogical research using brain localization as a prototype. We have discovered a core region in large language models that corresponds to linguistic competence. We observe that an improvement in linguistic competence does not necessarily accompany an elevation in the model's knowledge level.
  arXiv  Detail & Related papers  (2023-10-23T13:31:32Z)
• Journey to the Center of the Knowledge Neurons: Discoveries of Language-Independent Knowledge Neurons and Degenerate Knowledge Neurons [20.56154830853632]
  This paper delves into the complex task of understanding how factual knowledge is stored in multilingual language models. We introduce the Architecture-adapted Multilingual Integrated Gradients method, which successfully localizes knowledge neurons more precisely. We also conduct an in-depth exploration of knowledge neurons, leading to the following two important discoveries.
  arXiv  Detail & Related papers  (2023-08-25T06:26:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.