Beyond Similarity: A Gradient-based Graph Method for Instruction Tuning Data Selection
- URL: http://arxiv.org/abs/2502.11062v1
- Date: Sun, 16 Feb 2025 10:06:00 GMT
- Title: Beyond Similarity: A Gradient-based Graph Method for Instruction Tuning Data Selection
- Authors: Yang Zhao, Li Du, Xiao Ding, Yangou Ouyang, Hepeng Wang, Kai Xiong, Jinglong Gao, Zhouhao Sun, Dongliang Xu, Yang Qing, Dongchen Li, Bing Qin, Ting Liu,
- Abstract summary: Large language models (LLMs) have shown great potential across various industries due to their remarkable ability to generalize through instruction tuning.
We introduce G2IS (Gradient-based Graph Instruction Selection), a novel method that constructs a mixed gradient-based instruction graph to capture the joint distribution and interdependencies between instructions.
G2IS outperforms traditional methods across various domain adaptation tasks, yielding significant performance gains, particularly in complex, data-scarce scenarios.
- Score: 39.791180321058214
- License:
- Abstract: Large language models (LLMs) have shown great potential across various industries due to their remarkable ability to generalize through instruction tuning. However, the limited availability of domain-specific data significantly hampers their performance on specialized tasks. While existing methods primarily focus on selecting training data from general datasets that are similar to the target domain, they often fail to consider the joint distribution of instructions, resulting in inefficient learning and suboptimal knowledge transfer. To address these challenges, we introduce G2IS (Gradient-based Graph Instruction Selection), a novel method that constructs a mixed gradient-based instruction graph to capture the joint distribution and interdependencies between instructions. By accounting for the relationships between instructions, G2IS improves domain adaptation efficiency. Additionally, we propose a gradient walk algorithm to refine the data selection process, enhancing both training effectiveness and efficiency. Our experiments demonstrate that G2IS outperforms traditional methods across various domain adaptation tasks, yielding significant performance gains, particularly in complex, data-scarce scenarios. These results underscore the potential of G2IS in advancing the development of large, domain-specific models.
Related papers
- MLDGG: Meta-Learning for Domain Generalization on Graphs [9.872254367103057]
Domain generalization on graphs aims to develop models with robust generalization capabilities.
Our framework, MLDGG, endeavors to achieve adaptable generalization across diverse domains by integrating cross-multi-domain meta-learning.
Our empirical results demonstrate that MLDGG surpasses baseline methods, showcasing its effectiveness in three different distribution shift settings.
arXiv Detail & Related papers (2024-11-19T22:57:38Z) - Advancing Open-Set Domain Generalization Using Evidential Bi-Level Hardest Domain Scheduler [45.71475375161575]
In Open-Set Domain Generalization, the model is exposed to both new variations of data appearance (domains) and open-set conditions.
We propose the Evidential Bi-Level Hardest Domain Scheduler (EBiL-HaDS) to achieve an adaptive domain scheduler.
arXiv Detail & Related papers (2024-09-26T05:57:35Z) - LESS: Selecting Influential Data for Targeted Instruction Tuning [64.78894228923619]
We propose LESS, an efficient algorithm to estimate data influences and perform Low-rank gradiEnt Similarity Search for instruction data selection.
We show that training on a LESS-selected 5% of the data can often outperform training on the full dataset across diverse downstream tasks.
Our method goes beyond surface form cues to identify data that the necessary reasoning skills for the intended downstream application.
arXiv Detail & Related papers (2024-02-06T19:18:04Z) - Domain Adaptive Graph Classification [0.0]
We introduce the Dual Adversarial Graph Representation Learning (DAGRL), which explore the graph topology from dual branches and mitigate domain discrepancies via dual adversarial learning.
Our approach incorporates adaptive perturbations into the dual branches, which align the source and target distribution to address domain discrepancies.
arXiv Detail & Related papers (2023-12-21T02:37:56Z) - GraphControl: Adding Conditional Control to Universal Graph Pre-trained
Models for Graph Domain Transfer Learning [28.04023419006392]
Graph self-supervised algorithms have achieved significant success in acquiring generic knowledge from abundant unlabeled graph data.
Different graphs, even across seemingly similar domains, can differ significantly in terms of attribute semantics.
We introduce an innovative deployment module coined as GraphControl, motivated by ControlNet, to realize better graph domain transfer learning.
arXiv Detail & Related papers (2023-10-11T10:30:49Z) - Joint Optimization of Class-Specific Training- and Test-Time Data
Augmentation in Segmentation [35.41274775082237]
This paper presents an effective and general data augmentation framework for medical image segmentation.
We adopt a computationally efficient and data-efficient gradient-based meta-learning scheme to align the distribution of training and validation data.
We demonstrate the effectiveness of our method on four medical image segmentation tasks with two state-of-the-art segmentation models, DeepMedic and nnU-Net.
arXiv Detail & Related papers (2023-05-30T14:48:45Z) - Guiding Generative Language Models for Data Augmentation in Few-Shot
Text Classification [59.698811329287174]
We leverage GPT-2 for generating artificial training instances in order to improve classification performance.
Our results show that fine-tuning GPT-2 in a handful of label instances leads to consistent classification improvements.
arXiv Detail & Related papers (2021-11-17T12:10:03Z) - Domain Adaptive Person Re-Identification via Coupling Optimization [58.567492812339566]
Domain adaptive person Re-Identification (ReID) is challenging owing to the domain gap and shortage of annotations on target scenarios.
This paper proposes a coupling optimization method including the Domain-Invariant Mapping (DIM) method and the Global-Local distance Optimization ( GLO)
GLO is designed to train the ReID model with unsupervised setting on the target domain.
arXiv Detail & Related papers (2020-11-06T14:01:03Z) - Robust Optimization as Data Augmentation for Large-scale Graphs [117.2376815614148]
We propose FLAG (Free Large-scale Adversarial Augmentation on Graphs), which iteratively augments node features with gradient-based adversarial perturbations during training.
FLAG is a general-purpose approach for graph data, which universally works in node classification, link prediction, and graph classification tasks.
arXiv Detail & Related papers (2020-10-19T21:51:47Z) - Supervised Domain Adaptation using Graph Embedding [86.3361797111839]
Domain adaptation methods assume that distributions between the two domains are shifted and attempt to realign them.
We propose a generic framework based on graph embedding.
We show that the proposed approach leads to a powerful Domain Adaptation framework.
arXiv Detail & Related papers (2020-03-09T12:25:13Z)
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.