ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning

• URL: http://arxiv.org/abs/2412.00631v1
• Date: Sun, 01 Dec 2024 01:01:09 GMT
• Title: ROSE: A Reward-Oriented Data Selection Framework for LLM Task-Specific Instruction Tuning
• Authors: Yang Wu, Huayi Zhang, Yizheng Jiao, Lin Ma, Xiaozhong Liu, Jinhong Yu, Dongyu Zhang, Dezhi Yu, Wei Xu,
• Abstract summary: We introduce Reward-Oriented inStruction data sElection to optimize data selection for task-specific instruction tuning. ROSE adapts an influence formulation to approximate the influence of training data points relative to a few-shot preference validation set to select the most task-related training data points.
• Score: 29.001249598245
• License:
• Abstract: Instruction tuning has underscored the significant potential of large language models (LLMs) in producing more human-controllable and effective outputs in various domains. In this work, we focus on the data selection problem for task-specific instruction tuning of LLMs. Prevailing methods primarily rely on the crafted similarity metrics to select training data that aligns with the test data distribution. The goal is to minimize instruction tuning loss on the test data, ultimately improving performance on the target task. However, it has been widely observed that instruction tuning loss (i.e., cross-entropy loss for next token prediction) in LLMs often fails to exhibit a monotonic relationship with actual task performance. This misalignment undermines the effectiveness of current data selection methods for task-specific instruction tuning. To address this issue, we introduce ROSE, a novel Reward-Oriented inStruction data sElection method which leverages pairwise preference loss as a reward signal to optimize data selection for task-specific instruction tuning. Specifically, ROSE adapts an influence formulation to approximate the influence of training data points relative to a few-shot preference validation set to select the most task-related training data points. Experimental results show that by selecting just 5% of the training data using ROSE, our approach can achieve competitive results compared to fine-tuning with the full training dataset, and it surpasses other state-of-the-art data selection methods for task-specific instruction tuning. Our qualitative analysis further confirms the robust generalizability of our method across multiple benchmark datasets and diverse model architectures.

Related papers

• Aligning Instruction Tuning with Pre-training [81.4748965653345]
  We propose Aligning Instruction Tuning with Pre-training (AITP) to align instruction tuning with pre-training distributions. We show consistent performance improvements with AITP on three fully open large language models (LLMs) across eight benchmarks.
  arXiv  Detail & Related papers  (2025-01-16T08:27:40Z)
• Capturing the Temporal Dependence of Training Data Influence [100.91355498124527]
  We formalize the concept of trajectory-specific leave-one-out influence, which quantifies the impact of removing a data point during training. We propose data value embedding, a novel technique enabling efficient approximation of trajectory-specific LOO. As data value embedding captures training data ordering, it offers valuable insights into model training dynamics.
  arXiv  Detail & Related papers  (2024-12-12T18:28:55Z)
• TSDS: Data Selection for Task-Specific Model Finetuning [39.19448080265558]
  The efficacy of task-specific finetuning largely depends on the selection of appropriate training data. We present TSDS (Task-Specific Data Selection), a framework to select data for task-specific model finetuning. We show that instruction tuning using data selected by our method with a 1% selection ratio often outperforms using the full dataset.
  arXiv  Detail & Related papers  (2024-10-15T05:54:17Z)
• A CLIP-Powered Framework for Robust and Generalizable Data Selection [51.46695086779598]
  Real-world datasets often contain redundant and noisy data, imposing a negative impact on training efficiency and model performance. Data selection has shown promise in identifying the most representative samples from the entire dataset. We propose a novel CLIP-powered data selection framework that leverages multimodal information for more robust and generalizable sample selection.
  arXiv  Detail & Related papers  (2024-10-15T03:00:58Z)
• 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)
• A Survey on Data Selection for LLM Instruction Tuning [18.94987580516951]
  We propose a new taxonomy of the data selection methods and provide a detailed introduction of recent advances. We emphasize the open challenges and present new frontiers of this task.
  arXiv  Detail & Related papers  (2024-02-04T13:32:01Z)
• DsDm: Model-Aware Dataset Selection with Datamodels [81.01744199870043]
  Standard practice is to filter for examples that match human notions of data quality. We find that selecting according to similarity with "high quality" data sources may not increase (and can even hurt) performance compared to randomly selecting data. Our framework avoids handpicked notions of data quality, and instead models explicitly how the learning process uses train datapoints to predict on the target tasks.
  arXiv  Detail & Related papers  (2024-01-23T17:22:00Z)
• Improving Multi-Turn Response Selection Models with Complementary Last-Utterance Selection by Instance Weighting [84.9716460244444]
  We consider utilizing the underlying correlation in the data resource itself to derive different kinds of supervision signals. We conduct extensive experiments in two public datasets and obtain significant improvement in both datasets.
  arXiv  Detail & Related papers  (2020-02-18T06:29:01Z)

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.