Task-Adaptive Pretrained Language Models via Clustered-Importance Sampling

• URL: http://arxiv.org/abs/2410.03735v1
• Date: Mon, 30 Sep 2024 20:49:54 GMT
• Title: Task-Adaptive Pretrained Language Models via Clustered-Importance Sampling
• Authors: David Grangier, Simin Fan, Skyler Seto, Pierre Ablin,
• Abstract summary: We build specialist models from large generalist training sets instead. We adjust the training distribution of the generalist data with guidance from the limited domain-specific data. It is scalable, suitable for pretraining and continued pretraining, it works well in multi-task settings.
• Score: 21.762562172089236
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Specialist language models (LMs) focus on a specific task or domain on which they often outperform generalist LMs of the same size. However, the specialist data needed to pretrain these models is only available in limited amount for most tasks. In this work, we build specialist models from large generalist training sets instead. We adjust the training distribution of the generalist data with guidance from the limited domain-specific data. We explore several approaches, with clustered importance sampling standing out. This method clusters the generalist dataset and samples from these clusters based on their frequencies in the smaller specialist dataset. It is scalable, suitable for pretraining and continued pretraining, it works well in multi-task settings. Our findings demonstrate improvements across different domains in terms of language modeling perplexity and accuracy on multiple-choice question tasks. We also present ablation studies that examine the impact of dataset sizes, clustering configurations, and model sizes.

Related papers

• Adapt-$\infty$: Scalable Lifelong Multimodal Instruction Tuning via Dynamic Data Selection [89.42023974249122]
  Adapt-$infty$ is a new multi-way and adaptive data selection approach for Lifelong Instruction Tuning. We construct pseudo-skill clusters by grouping gradient-based sample vectors. We select the best-performing data selector for each skill cluster from a pool of selector experts.
  arXiv  Detail & Related papers  (2024-10-14T15:48:09Z)
• NuwaTS: a Foundation Model Mending Every Incomplete Time Series [24.768755438620666]
  We present textbfNuwaTS, a novel framework that repurposes Pre-trained Language Models for general time series imputation. NuwaTS can be applied to impute missing data across any domain. We show that NuwaTS generalizes to other time series tasks, such as forecasting.
  arXiv  Detail & Related papers  (2024-05-24T07:59:02Z)
• Do Membership Inference Attacks Work on Large Language Models? [141.2019867466968]
  Membership inference attacks (MIAs) attempt to predict whether a particular datapoint is a member of a target model's training data. We perform a large-scale evaluation of MIAs over a suite of language models trained on the Pile, ranging from 160M to 12B parameters. We find that MIAs barely outperform random guessing for most settings across varying LLM sizes and domains.
  arXiv  Detail & Related papers  (2024-02-12T17:52:05Z)
• Task-customized Masked AutoEncoder via Mixture of Cluster-conditional Experts [104.9871176044644]
  Masked Autoencoder(MAE) is a prevailing self-supervised learning method that achieves promising results in model pre-training. We propose a novel MAE-based pre-training paradigm, Mixture of Cluster-conditional Experts (MoCE) MoCE trains each expert only with semantically relevant images by using cluster-conditional gates.
  arXiv  Detail & Related papers  (2024-02-08T03:46:32Z)
• Large Pre-trained time series models for cross-domain Time series analysis tasks [20.228846068418765]
  We propose a novel method of textitadaptive segmentation that automatically identifies optimal dataset-specific segmentation strategy during pre-training. This enables LPTM to perform similar to or better than domain-specific state-of-art model when fine-tuned to different downstream time-series analysis tasks and under zero-shot settings.
  arXiv  Detail & Related papers  (2023-11-19T20:16:16Z)
• Unsupervised Calibration through Prior Adaptation for Text Classification using Large Language Models [37.39843935632105]
  We propose an approach to adapt the prior class distribution to perform text classification tasks without the need for labelled samples. Results show that these methods outperform the un-adapted model for different number of training shots in the prompt.
  arXiv  Detail & Related papers  (2023-07-13T12:11:36Z)
• Scaling Expert Language Models with Unsupervised Domain Discovery [107.08940500543447]
  We introduce a simple but effective method to asynchronously train large, sparse language models on arbitrary text corpora. Our method clusters a corpus into sets of related documents, trains a separate expert language model on each cluster, and combines them in a sparse ensemble for inference.
  arXiv  Detail & Related papers  (2023-03-24T17:38:58Z)
• Meta-learning Pathologies from Radiology Reports using Variance Aware Prototypical Networks [3.464871689508835]
  We propose a simple extension of the Prototypical Networks for few-shot text classification. Our main idea is to replace the class prototypes by Gaussians and introduce a regularization term that encourages the examples to be clustered near the appropriate class centroids.
  arXiv  Detail & Related papers  (2022-10-22T05:22:29Z)
• CHALLENGER: Training with Attribution Maps [63.736435657236505]
  We show that utilizing attribution maps for training neural networks can improve regularization of models and thus increase performance. In particular, we show that our generic domain-independent approach yields state-of-the-art results in vision, natural language processing and on time series tasks.
  arXiv  Detail & Related papers  (2022-05-30T13:34:46Z)
• A Generative Language Model for Few-shot Aspect-Based Sentiment Analysis [90.24921443175514]
  We focus on aspect-based sentiment analysis, which involves extracting aspect term, category, and predicting their corresponding polarities. We propose to reformulate the extraction and prediction tasks into the sequence generation task, using a generative language model with unidirectional attention. Our approach outperforms the previous state-of-the-art (based on BERT) on average performance by a large margins in few-shot and full-shot settings.
  arXiv  Detail & Related papers  (2022-04-11T18:31:53Z)
• Zero-shot meta-learning for small-scale data from human subjects [10.320654885121346]
  We develop a framework to rapidly adapt to a new prediction task with limited training data for out-of-sample test data. Our model learns the latent treatment effects of each intervention and, by design, can naturally handle multi-task predictions. Our model has implications for improved generalization of small-size human studies to the wider population.
  arXiv  Detail & Related papers  (2022-03-29T17:42:04Z)

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.