Pre-training Distillation for Large Language Models: A Design Space Exploration

• URL: http://arxiv.org/abs/2410.16215v1
• Date: Mon, 21 Oct 2024 17:16:13 GMT
• Title: Pre-training Distillation for Large Language Models: A Design Space Exploration
• Authors: Hao Peng, Xin Lv, Yushi Bai, Zijun Yao, Jiajie Zhang, Lei Hou, Juanzi Li,
• Abstract summary: Pre-training distillation aims to transfer knowledge from a large teacher model to a smaller student model. We conduct experiments to explore the design space of pre-training distillation and find better configurations. We hope our exploration of the design space will inform future practices in pre-training distillation.
• Score: 54.67324039434781
• License:
• Abstract: Knowledge distillation (KD) aims to transfer knowledge from a large teacher model to a smaller student model. Previous work applying KD in the field of large language models (LLMs) typically focused on the post-training phase, where the student LLM learns directly from instructions and corresponding responses generated by the teacher model. In this paper, we extend KD to the pre-training phase of LLMs, named pre-training distillation (PD). We first conduct a preliminary experiment using GLM-4-9B as the teacher LLM to distill a 1.9B parameter student LLM, validating the effectiveness of PD. Considering the key impact factors of distillation, we systematically explore the design space of pre-training distillation across four aspects: logits processing, loss selection, scaling law, and offline or online logits. We conduct extensive experiments to explore the design space of pre-training distillation and find better configurations and interesting conclusions, such as larger student LLMs generally benefiting more from pre-training distillation, while a larger teacher LLM does not necessarily guarantee better results. We hope our exploration of the design space will inform future practices in pre-training distillation.

Related papers

• LLaVA-MoD: Making LLaVA Tiny via MoE Knowledge Distillation [41.05687297326706]
  LLaVA-MoD is a framework designed to enable the efficient training of small-scale Multimodal Language Models. We optimize the network structure of s-MLLM by integrating a sparse Mixture of Experts architecture into the language model. We also propose a progressive knowledge transfer strategy to ensure comprehensive knowledge migration.
  arXiv  Detail & Related papers  (2024-08-28T15:52:23Z)
• LLAVADI: What Matters For Multimodal Large Language Models Distillation [77.73964744238519]
  In this work, we do not propose a new efficient model structure or train small-scale MLLMs from scratch. Our studies involve training strategies, model choices, and distillation algorithms in the knowledge distillation process. By evaluating different benchmarks and proper strategy, even a 2.7B small-scale model can perform on par with larger models with 7B or 13B parameters.
  arXiv  Detail & Related papers  (2024-07-28T06:10:47Z)
• Direct Preference Knowledge Distillation for Large Language Models [73.50849692633953]
  We propose Direct Preference Knowledge Distillation (DPKD) for large language models (LLMs) We re-formulate KD of LLMs into two stages: first optimizing and objective consisting of implicit reward and reverse KL divergence. We prove the value and effectiveness of the introduced implicit reward and output preference in KD through experiments and theoretical analysis.
  arXiv  Detail & Related papers  (2024-06-28T09:23:40Z)
• PLaD: Preference-based Large Language Model Distillation with Pseudo-Preference Pairs [47.35598271306371]
  Large Language Models (LLMs) have exhibited impressive capabilities in various tasks, yet their vast parameter sizes restrict their applicability in resource-constrained settings. Knowledge distillation (KD) offers a viable solution by transferring expertise from large teacher models to compact student models. We present PLaD, a novel preference-based LLM distillation framework.
  arXiv  Detail & Related papers  (2024-06-05T03:08:25Z)
• Towards Optimal Learning of Language Models [124.65669486710992]
  We present a theory for the optimal learning of language models (LMs) We derive a theorem, named Learning Law, to reveal the properties of the dynamics in the optimal learning process under our objective. We empirically verify that the optimal learning of LMs essentially stems from the improvement of the coefficients in the scaling law of LMs.
  arXiv  Detail & Related papers  (2024-02-27T18:52:19Z)
• ELAD: Explanation-Guided Large Language Models Active Distillation [16.243249111524403]
  The deployment and application of Large Language Models (LLMs) is hindered by their memory inefficiency, computational demands, and the high costs of API inferences. Traditional distillation methods, which transfer the capabilities of LLMs to smaller models, often fail to determine whether the knowledge has been sufficiently transferred. We propose an Explanation-Guided LLMs Active Distillation (ELAD) framework that employs an active learning strategy to optimize the balance between annotation costs and model performance.
  arXiv  Detail & Related papers  (2024-02-20T15:47:59Z)
• Knowledge Inheritance for Pre-trained Language Models [57.51305807391381]
  We introduce a novel pre-training framework named "knowledge inheritance" (KI) KI combines both self-learning and teacher-guided learning to efficiently train larger PLMs. We show that KI can well support lifelong learning and knowledge transfer.
  arXiv  Detail & Related papers  (2021-05-28T14:43:26Z)

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.