How Do Large Language Models Learn Concepts During Continual Pre-Training?
Human beings primarily understand the world through concepts (e.g., dog), abstract mental representations that structure perception, reasoning, and learning. However, how large language models (LLMs) acquire, retain, and forget such concepts during continual pretraining remains poorly understood. In this work, we study how individual concepts are acquired and forgotten, as well as how multiple concepts interact through interference and synergy. We link these behavioral dynamics to LLMs' internal Concept Circuits, computational subgraphs associated with specific concepts, and incorporate Graph Metrics to characterize circuit structure. Our analysis reveals: (1) LLMs concept circuits provide a non-trivial, statistically significant signal of concept learning and forgetting; (2) Concept circuits exhibit a stage-wise temporal pattern during continual pretraining, with an early increase followed by gradual decrease and stabilization; (3) concepts with larger learning gains tend to exhibit greater forgetting under subsequent training; (4) semantically similar concepts induce stronger interference than weakly related ones; (5) conceptual knowledge differs in their transferability, with some significantly facilitating the learning of others. Together, our findings offer a circuit-level view of concept learning dynamics and inform the design of more interpretable and robust concept-aware training strategies for LLMs.
Code (0)
등록된 구현이 없습니다.
Tasks
Continual PretrainingSimilar Papers 제목 키워드 기반
Continually Learn to Map Visual Concepts to Large Language Models in Resource-constrained Environments
Learning continually from a stream of non-i.i.d. data is an open challenge in deep learning, even more so when working in resource-constrained environments such as embedded devices. Visual models that are continually upd…
Continual LearningLanguage ModelingLanguage ModellingLarge Language ModelCLeAR: Continual Learning on Algorithmic Reasoning for Human-like Intelligence
Continual learning (CL) aims to incrementally learn multiple tasks that are presented sequentially. The significance of CL lies not only in the practical importance but also in studying the learning mechanisms of humans …
Which Concepts to Forget and How to Refuse? Decomposing Concepts for Continual Unlearning in Large Vision-Language Models
Continual unlearning poses the challenge of enabling large vision-language models to selectively refuse specific image-instruction pairs in response to sequential deletion requests, while preserving general utility. Howe…
Incremental Concept Learning via Online Generative Memory Recall
The ability to learn more and more concepts over time from incrementally arriving data is essential for the development of a life-long learning system. However, deep neural networks often suffer from forgetting previousl…
class-incremental learningClass Incremental LearningGenerative Adversarial NetworkIncremental LearningAugmenting Continual Learning of Diseases with LLM-Generated Visual Concepts
Continual learning is essential for medical image classification systems to adapt to dynamically evolving clinical environments. The integration of multimodal information can significantly enhance continual learning of i…
Medical Image ClassificationContinual Learning