Breaking the Stage Barrier: A Novel Single-Stage Approach to Long Context Extension for Large Language Models

• URL: http://arxiv.org/abs/2412.07171v1
• Date: Tue, 10 Dec 2024 04:09:29 GMT
• Title: Breaking the Stage Barrier: A Novel Single-Stage Approach to Long Context Extension for Large Language Models
• Authors: Haoran Lian, Junmin Chen, Wei Huang, Yizhe Xiong, Wenping Hu, Guiguang Ding, Hui Chen, Jianwei Niu, Zijia Lin, Fuzheng Zhang, Di Zhang,
• Abstract summary: Large language models (LLMs) have revolutionized Natural Language Processing (NLP) Currently solutions toward long context modeling often employ multi-stage continual pertaining. In this paper, we introduce a novel single-stage continual pretraining method, Head-Adaptive Rotary Position.
• Score: 28.253786579346432
• License:
• Abstract: Recently, Large language models (LLMs) have revolutionized Natural Language Processing (NLP). Pretrained LLMs, due to limited training context size, struggle with handling long token sequences, limiting their performance on various downstream tasks. Current solutions toward long context modeling often employ multi-stage continual pertaining, which progressively increases the effective context length through several continual pretraining stages. However, those approaches require extensive manual tuning and human expertise. In this paper, we introduce a novel single-stage continual pretraining method, Head-Adaptive Rotary Position Encoding (HARPE), to equip LLMs with long context modeling capabilities while simplifying the training process. Our HARPE leverages different Rotary Position Encoding (RoPE) base frequency values across different attention heads and directly trains LLMs on the target context length. Extensive experiments on 4 language modeling benchmarks, including the latest RULER benchmark, demonstrate that HARPE excels in understanding and integrating long-context tasks with single-stage training, matching and even outperforming existing multi-stage methods. Our results highlight that HARPE successfully breaks the stage barrier for training LLMs with long context modeling capabilities.

Related papers

• Scalable Language Model with Generalized Continual Learning [58.700439919096155]
  The Joint Adaptive Re-ization (JARe) is integrated with Dynamic Task-related Knowledge Retrieval (DTKR) to enable adaptive adjustment of language models based on specific downstream tasks. Our method demonstrates state-of-the-art performance on diverse backbones and benchmarks, achieving effective continual learning in both full-set and few-shot scenarios with minimal forgetting.
  arXiv  Detail & Related papers  (2024-04-11T04:22:15Z)
• Continual Learning for Large Language Models: A Survey [95.79977915131145]
  Large language models (LLMs) are not amenable to frequent re-training, due to high training costs arising from their massive scale. This paper surveys recent works on continual learning for LLMs.
  arXiv  Detail & Related papers  (2024-02-02T12:34:09Z)
• E^2-LLM: Efficient and Extreme Length Extension of Large Language Models [74.1254067728251]
  We propose an Efficient and Extreme length extension method for Large Language Models, called E 2 -LLM, with only one training procedure and dramatically reduced cost. Comprehensive experimental results on multiple benchmark datasets demonstrate the effectiveness of our E 2 -LLM on challenging long-context tasks.
  arXiv  Detail & Related papers  (2024-01-13T02:11:20Z)
• ModaVerse: Efficiently Transforming Modalities with LLMs [25.49713745405194]
  We introduce ModaVerse, a Multi-modal Large Language Model capable of comprehending and transforming content across various modalities. We propose a novel Input/Output (I/O) alignment mechanism that operates directly at the level of natural language.
  arXiv  Detail & Related papers  (2024-01-12T06:28:54Z)
• CLEX: Continuous Length Extrapolation for Large Language Models [68.43814043853347]
  We propose Continuous Length EXtrapolation (CLEX) for Large Language Models (LLMs) CLEX extends the context window to over 4x or almost 8x training length, with no deterioration in performance. Our model trained on a 4k length exhibits competitive performance against state-of-the-art open-source models trained on context lengths up to 32k.
  arXiv  Detail & Related papers  (2023-10-25T08:13:02Z)
• GrowLength: Accelerating LLMs Pretraining by Progressively Growing Training Length [65.24730341801468]
  This paper introduces a novel, simple, and effective method named growlength'' to accelerate the pretraining process of Large Language Models. Our method progressively increases the training length throughout the pretraining phase, thereby mitigating computational costs and enhancing efficiency.
  arXiv  Detail & Related papers  (2023-10-01T05:25:24Z)
• Effective Long-Context Scaling of Foundation Models [90.57254298730923]
  We present a series of long-context LLMs that support effective context windows of up to 32,768 tokens. Our models achieve consistent improvements on most regular tasks and significant improvements on long-context tasks over Llama 2.
  arXiv  Detail & Related papers  (2023-09-27T21:41:49Z)

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.