Zero-Shot Generalization during Instruction Tuning: Insights from Similarity and Granularity

• URL: http://arxiv.org/abs/2406.11721v1
• Date: Mon, 17 Jun 2024 16:40:21 GMT
• Title: Zero-Shot Generalization during Instruction Tuning: Insights from Similarity and Granularity
• Authors: Bingxiang He, Ning Ding, Cheng Qian, Jia Deng, Ganqu Cui, Lifan Yuan, Huan-ang Gao, Huimin Chen, Zhiyuan Liu, Maosong Sun,
• Abstract summary: We show that zero-shot generalization during instruction tuning happens very early. We also show that encountering highly similar and fine-grained training data earlier during instruction tuning, without the constraints of defined "tasks", enables better generalization. For the first time, we show that zero-shot generalization during instruction tuning is a form of similarity-based generalization between training and test data at the instance level.
• Score: 84.12126298229866
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding alignment techniques begins with comprehending zero-shot generalization brought by instruction tuning, but little of the mechanism has been understood. Existing work has largely been confined to the task level, without considering that tasks are artificially defined and, to LLMs, merely consist of tokens and representations. This line of research has been limited to examining transfer between tasks from a task-pair perspective, with few studies focusing on understanding zero-shot generalization from the perspective of the data itself. To bridge this gap, we first demonstrate through multiple metrics that zero-shot generalization during instruction tuning happens very early. Next, we investigate the facilitation of zero-shot generalization from both data similarity and granularity perspectives, confirming that encountering highly similar and fine-grained training data earlier during instruction tuning, without the constraints of defined "tasks", enables better generalization. Finally, we propose a more grounded training data arrangement method, Test-centric Multi-turn Arrangement, and show its effectiveness in promoting continual learning and further loss reduction. For the first time, we show that zero-shot generalization during instruction tuning is a form of similarity-based generalization between training and test data at the instance level. We hope our analysis will advance the understanding of zero-shot generalization during instruction tuning and contribute to the development of more aligned LLMs. Our code is released at https://github.com/HBX-hbx/dynamics_of_zero-shot_generalization.

Related papers

• Mind the Interference: Retaining Pre-trained Knowledge in Parameter Efficient Continual Learning of Vision-Language Models [79.28821338925947]
  Domain-Class Incremental Learning is a realistic but challenging continual learning scenario. To handle these diverse tasks, pre-trained Vision-Language Models (VLMs) are introduced for their strong generalizability. This incurs a new problem: the knowledge encoded in the pre-trained VLMs may be disturbed when adapting to new tasks, compromising their inherent zero-shot ability. Existing methods tackle it by tuning VLMs with knowledge distillation on extra datasets, which demands heavy overhead. We propose the Distribution-aware Interference-free Knowledge Integration (DIKI) framework, retaining pre-trained knowledge of
  arXiv  Detail & Related papers  (2024-07-07T12:19:37Z)
• Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation [14.225723195634941]
  We propose a novel approach to prompt learning based on unsupervised knowledge distillation from more powerful models. Our approach, which we call Knowledge Distillation Prompt Learning (KDPL), can be integrated into existing prompt learning techniques.
  arXiv  Detail & Related papers  (2024-07-03T12:24:40Z)
• On the Generalization Ability of Unsupervised Pretraining [53.06175754026037]
  Recent advances in unsupervised learning have shown that unsupervised pre-training, followed by fine-tuning, can improve model generalization. This paper introduces a novel theoretical framework that illuminates the critical factor influencing the transferability of knowledge acquired during unsupervised pre-training to the subsequent fine-tuning phase. Our results contribute to a better understanding of unsupervised pre-training and fine-tuning paradigm, and can shed light on the design of more effective pre-training algorithms.
  arXiv  Detail & Related papers  (2024-03-11T16:23:42Z)
• Closing the Gap between TD Learning and Supervised Learning -- A Generalisation Point of View [51.30152184507165]
  Some reinforcement learning (RL) algorithms can stitch pieces of experience to solve a task never seen before during training. This oft-sought property is one of the few ways in which RL methods based on dynamic-programming differ from RL methods based on supervised-learning (SL) It remains unclear whether those methods forgo this important stitching property.
  arXiv  Detail & Related papers  (2024-01-20T14:23:25Z)
• Hierarchical Decomposition of Prompt-Based Continual Learning: Rethinking Obscured Sub-optimality [55.88910947643436]
  Self-supervised pre-training is essential for handling vast quantities of unlabeled data in practice. HiDe-Prompt is an innovative approach that explicitly optimize the hierarchical components with an ensemble of task-specific prompts and statistics. Our experiments demonstrate the superior performance of HiDe-Prompt and its robustness to pre-training paradigms in continual learning.
  arXiv  Detail & Related papers  (2023-10-11T06:51:46Z)
• Instruction Position Matters in Sequence Generation with Large Language Models [67.87516654892343]
  Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization. We propose enhancing the instruction-following capability of LLMs by shifting the position of task instructions after the input sentences.
  arXiv  Detail & Related papers  (2023-08-23T12:36:57Z)
• Towards the Generalization of Contrastive Self-Supervised Learning [11.889992921445849]
  We present a theoretical explanation of how contrastive self-supervised pre-trained models generalize to downstream tasks. We further explore SimCLR and Barlow Twins, which are two canonical contrastive self-supervised methods.
  arXiv  Detail & Related papers  (2021-11-01T07:39:38Z)
• Explaining generalization in deep learning: progress and fundamental limits [8.299945169799795]
  In the first part of the thesis, we will empirically study how training deep networks via gradient descent implicitly controls the networks' capacity. We will then derive em data-dependent em uniform-convergence-based generalization bounds with improved dependencies on the parameter count. In the last part of the thesis, we will introduce an em empirical technique to estimate generalization using unlabeled data.
  arXiv  Detail & Related papers  (2021-10-17T21:17:30Z)

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.