Exploring Forgetting in Large Language Model Pre-Training

• URL: http://arxiv.org/abs/2410.17018v1
• Date: Tue, 22 Oct 2024 13:39:47 GMT
• Title: Exploring Forgetting in Large Language Model Pre-Training
• Authors: Chonghua Liao, Ruobing Xie, Xingwu Sun, Haowen Sun, Zhanhui Kang,
• Abstract summary: Catastrophic forgetting remains a formidable obstacle to building an omniscient model in large language models (LLMs) We systematically explored the existence and measurement of forgetting in pre-training, questioning traditional metrics such as perplexity (PPL) and introducing new metrics to better detect entity memory retention.
• Score: 18.858330348834777
• License:
• Abstract: Catastrophic forgetting remains a formidable obstacle to building an omniscient model in large language models (LLMs). Despite the pioneering research on task-level forgetting in LLM fine-tuning, there is scant focus on forgetting during pre-training. We systematically explored the existence and measurement of forgetting in pre-training, questioning traditional metrics such as perplexity (PPL) and introducing new metrics to better detect entity memory retention. Based on our revised assessment of forgetting metrics, we explored low-cost, straightforward methods to mitigate forgetting during the pre-training phase. Further, we carefully analyzed the learning curves, offering insights into the dynamics of forgetting. Extensive evaluations and analyses on forgetting of pre-training could facilitate future research on LLMs.

Related papers

• The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws [51.608402959163925]
  We present the first systematic exploration of optimal sparse pre-training configurations for large language models. We find that initiating pruning at 25% of total training compute and concluding at 75% achieves near-optimal final evaluation loss. We propose a new scaling law that modifies the Chinchilla scaling law to use the average parameter count over pre-training.
  arXiv  Detail & Related papers  (2025-01-21T20:23:22Z)
• Adaptive Rank, Reduced Forgetting: Knowledge Retention in Continual Learning Vision-Language Models with Dynamic Rank-Selective LoRA [19.982853959240497]
  Existing methods often rely on additional reference data, isolated components for distribution or domain predictions. We propose Dynamic Rank-Selective Low Rank Adaptation (LoRA), a universal and efficient continual learning approach. Our approach continually enhances the pre-trained VLM by retaining both the pre-trained knowledge and the knowledge acquired during CL.
  arXiv  Detail & Related papers  (2024-12-01T23:41:42Z)
• A Practical Guide to Fine-tuning Language Models with Limited Data [9.413178499853156]
  Employing pre-trained Large Language Models (LLMs) has become the de facto standard in Natural Language Processing (NLP) despite their extensive data requirements. Motivated by the recent surge in research focused on training LLMs with limited data, this paper surveys recent transfer learning approaches to optimize model performance in downstream tasks where data is scarce.
  arXiv  Detail & Related papers  (2024-11-14T15:55:37Z)
• Dynamic Uncertainty Ranking: Enhancing Retrieval-Augmented In-Context Learning for Long-Tail Knowledge in LLMs [50.29035873837]
  Large language models (LLMs) can learn vast amounts of knowledge from diverse domains during pre-training. Long-tail knowledge from specialized domains is often scarce and underrepresented, rarely appearing in the models' memorization. We propose a reinforcement learning-based dynamic uncertainty ranking method for ICL that accounts for the varying impact of each retrieved sample on LLM predictions.
  arXiv  Detail & Related papers  (2024-10-31T03:42:17Z)
• Deciphering Cross-Modal Alignment in Large Vision-Language Models with Modality Integration Rate [118.37653302885607]
  We present the Modality Integration Rate (MIR), an effective, robust, and generalized metric to indicate the multi-modal pre-training quality of Large Vision Language Models (LVLMs) MIR is indicative about training data selection, training strategy schedule, and model architecture design to get better pre-training results.
  arXiv  Detail & Related papers  (2024-10-09T17:59:04Z)
• Breaking Language Barriers: Cross-Lingual Continual Pre-Training at Scale [18.015805664219673]
  We explore an alternative approach to constructing an Large Language Model by continually pretraining (CPT) from existing pretrained LLMs. We find that CPT converges faster and saves significant resources in a scalable manner. The effectiveness of transfer at scale is influenced by training duration and linguistic properties, while robust to data replaying.
  arXiv  Detail & Related papers  (2024-07-02T10:06:41Z)
• The Fine Line: Navigating Large Language Model Pretraining with Down-streaming Capability Analysis [27.310894780313618]
  This paper undertakes a comprehensive comparison of model capabilities at various pretraining intermediate checkpoints. We confirm that specific downstream metrics exhibit similar training dynamics across models of different sizes. In addition to our core findings, we've reproduced Amber and OpenLLaMA, releasing their intermediate checkpoints.
  arXiv  Detail & Related papers  (2024-04-01T16:00:01Z)
• Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models [47.439995799065755]
  We pioneer the exploration of LLM's trustworthiness during pre-training. We focus on five key dimensions: reliability, privacy, toxicity, fairness, and robustness. We are the first to observe a similar two-phase phenomenon: fitting and compression.
  arXiv  Detail & Related papers  (2024-02-29T18:55:06Z)
• Few-Shot Class-Incremental Learning with Prior Knowledge [94.95569068211195]
  We propose Learning with Prior Knowledge (LwPK) to enhance the generalization ability of the pre-trained model. Experimental results indicate that LwPK effectively enhances the model resilience against catastrophic forgetting.
  arXiv  Detail & Related papers  (2024-02-02T08:05:35Z)
• Examining Forgetting in Continual Pre-training of Aligned Large Language Models [66.62800021628276]
  We investigate the phenomenon of forgetting that occurs during continual pre-training on an existing fine-tuned LLM. Experiment results highlight the non-trivial challenge of addressing catastrophic forgetting during continual pre-training.
  arXiv  Detail & Related papers  (2024-01-06T05:34:09Z)
• Knowledge Editing for Large Language Models: A Survey [51.01368551235289]
  One major drawback of large language models (LLMs) is their substantial computational cost for pre-training. Knowledge-based Model Editing (KME) has attracted increasing attention, which aims to precisely modify the LLMs to incorporate specific knowledge.
  arXiv  Detail & Related papers  (2023-10-24T22:18:13Z)

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.