ICML2021 | Self-Tuning: 如何减少对标记数据的需求？ - 知乎
本文介绍ICML2021的中稿论文：Self-Tuning for Data-Efficient Deep Learning，就“如何减少对标记数据的需求”这一重要问题给出了我们的思考。 大规模标记数据集推动深度学习获得了广泛应用，然而，在现实场景中收集足量的标记数据往往耗时耗力。 为了减少对标记数据的需求，半监督学习和迁移学习的研究者们从两个不同的视角给出了自己的思考：半监督学习 (Semi-supervised Learning, SSL)侧重于同时探索标记数据和无标记数据， 这里是清华大学软件学院机器学习组 …
大语言模型的指令微调（Instruction Tuning）最全综述：从数据集 …
2023年8月27日 · 微调（Fine-tuning）是一种在自然语言处理（NLP）中使用的技术，用于将预训练的语言模型适应于特定任务或领域。Fine-tuning的基本思想是采用已经在大量文本上进行训练的预训练语言模型，然后在小规模的任务特定文本上
How to fine-tune: Focus on effective datasets - Meta AI
2024年8月7日 · In this post, we explore some rules of thumb for curating a good training dataset. In Part 1, we took a look at prevalent approaches for adapting language models to domain data. In Part 2, we discussed how to determine if fine-tuning is the right approach for your use case.
SFT微调的数据组合及训练策略如何影响大模型性能：4个经典问题 …
2023年10月26日 · 本文主要介绍了 《how abilities in large language models are affected by supervised fine-tuning data com- position》 这一工作， 该工作探讨了sft阶段的数据组成，重点是数学推理、代码生成和一般的人工对齐能力。
Deep Learning Tuning Playbook - GitHub
Summary: for logging, evals, RNGs, checkpointing, and data sharding, multi-host training can make it very easy to introduce bugs! Ensure the pipeline is only logging and checkpointing on one host. Make sure before evaluation or checkpointing is run, the batch norm statistics are synchronized across hosts.
Introduction to tuning | Generative AI on Vertex AI - Google Cloud
6 天之前 · Model tuning is a crucial process in adapting Gemini to perform specific tasks with greater precision and accuracy. Model tuning works by providing a model with a training dataset that...
What Makes Good Data for Alignment? A Comprehensive Study …
2023年12月25日 · Recent research indicates the critical role of data engineering in instruction tuning -- when appropriately selected, only limited data is necessary to achieve superior performance. However, we still lack a principled understanding of what makes good instruction tuning data for alignment, and how we should select data automatically and effectively.
Reflection-Tuning: Data Recycling Improves LLM Instruction-Tuning
2023年10月18日 · We propose a novel method, termed "reflection-tuning," which addresses the problem by self-improvement and judging capabilities of LLMs. This approach utilizes an oracle LLM to recycle the original training data by introspecting and enhancing the quality of instructions and responses in the data.
[2102.12903] Self-Tuning for Data-Efficient Deep Learning
2021年2月25日 · To escape from this dilemma, we present Self-Tuning to enable data-efficient deep learning by unifying the exploration of labeled and unlabeled data and the transfer of a pre-trained model, as well as a Pseudo Group Contrast (PGC) mechanism to mitigate the reliance on pseudo-labels and boost the tolerance to false labels.
A Survey on Data-driven Performance Tuning for Big Data …
2021年7月15日 · We survey and describe data-driven tuning solutions for big data and present them in an organized and categorized way. We also describe tuning advisors and self-tuning proposals for big data environments, extract general and system-specific tuning recommendations and discuss research opportunities.
- 某些结果已被删除