Make Every Example Count: On the Stability and Utility of Self-Influence
for Learning from Noisy NLP Datasets
- URL: http://arxiv.org/abs/2302.13959v2
- Date: Tue, 17 Oct 2023 16:03:05 GMT
- Title: Make Every Example Count: On the Stability and Utility of Self-Influence
for Learning from Noisy NLP Datasets
- Authors: Irina Bejan, Artem Sokolov, Katja Filippova
- Abstract summary: We study the fitness of task-agnostic self-influence scores of training examples for data cleaning.
We analyze their efficacy in capturing naturally occurring outliers.
- Score: 4.142507103595571
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Increasingly larger datasets have become a standard ingredient to advancing
the state-of-the-art in NLP. However, data quality might have already become
the bottleneck to unlock further gains. Given the diversity and the sizes of
modern datasets, standard data filtering is not straight-forward to apply,
because of the multifacetedness of the harmful data and elusiveness of
filtering rules that would generalize across multiple tasks. We study the
fitness of task-agnostic self-influence scores of training examples for data
cleaning, analyze their efficacy in capturing naturally occurring outliers, and
investigate to what extent self-influence based data cleaning can improve
downstream performance in machine translation, question answering and text
classification, building up on recent approaches to self-influence calculation
and automated curriculum learning.
Related papers
- 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) - Self-Evolved Diverse Data Sampling for Efficient Instruction Tuning [47.02160072880698]
We introduce a self-evolving mechanism that allows the model itself to actively sample subsets that are equally or even more effective.
The key to our data sampling technique lies in the enhancement of diversity in the chosen subsets.
Extensive experiments across three datasets and benchmarks demonstrate the effectiveness of DiverseEvol.
arXiv Detail & Related papers (2023-11-14T14:10:40Z) - When Less is More: Investigating Data Pruning for Pretraining LLMs at
Scale [12.94829977468838]
Large volumes of text data have contributed significantly to the development of large language models.
To date, efforts to prune datasets down to a higher quality subset have relied on hand-crafteds encoded as rule-based filters.
We take a wider view and explore scalable estimates of data quality that can be used to measure the quality of pretraining data.
arXiv Detail & Related papers (2023-09-08T19:34:05Z) - A Pretrainer's Guide to Training Data: Measuring the Effects of Data
Age, Domain Coverage, Quality, & Toxicity [84.6421260559093]
This study is the largest set of experiments to validate, quantify, and expose undocumented intuitions about text pretraining.
Our findings indicate there does not exist a one-size-fits-all solution to filtering training data.
arXiv Detail & Related papers (2023-05-22T15:57:53Z) - Self-supervised Activity Representation Learning with Incremental Data:
An Empirical Study [7.782045150068569]
This research examines the impact of using a self-supervised representation learning model for time series classification tasks.
We analyzed the effect of varying the size, distribution, and source of the unlabeled data on the final classification performance across four public datasets.
arXiv Detail & Related papers (2023-05-01T01:39:55Z) - On-the-fly Denoising for Data Augmentation in Natural Language
Understanding [101.46848743193358]
We propose an on-the-fly denoising technique for data augmentation that learns from soft augmented labels provided by an organic teacher model trained on the cleaner original data.
Our method can be applied to general augmentation techniques and consistently improve the performance on both text classification and question-answering tasks.
arXiv Detail & Related papers (2022-12-20T18:58:33Z) - Training Dynamic based data filtering may not work for NLP datasets [0.0]
We study the applicability of the Area Under the Margin (AUM) metric to identify mislabelled examples in NLP datasets.
We find that mislabelled samples can be filtered using the AUM metric in NLP datasets but it also removes a significant number of correctly labeled points.
arXiv Detail & Related papers (2021-09-19T18:50:45Z) - Out-distribution aware Self-training in an Open World Setting [62.19882458285749]
We leverage unlabeled data in an open world setting to further improve prediction performance.
We introduce out-distribution aware self-training, which includes a careful sample selection strategy.
Our classifiers are by design out-distribution aware and can thus distinguish task-related inputs from unrelated ones.
arXiv Detail & Related papers (2020-12-21T12:25:04Z) - Omni-supervised Facial Expression Recognition via Distilled Data [120.11782405714234]
We propose omni-supervised learning to exploit reliable samples in a large amount of unlabeled data for network training.
We experimentally verify that the new dataset can significantly improve the ability of the learned FER model.
To tackle this, we propose to apply a dataset distillation strategy to compress the created dataset into several informative class-wise images.
arXiv Detail & Related papers (2020-05-18T09:36:51Z)
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.