BERT Goes Off-Topic: Investigating the Domain Transfer Challenge using Genre Classification

• URL: http://arxiv.org/abs/2311.16083v1
• Date: Mon, 27 Nov 2023 18:53:31 GMT
• Title: BERT Goes Off-Topic: Investigating the Domain Transfer Challenge using Genre Classification
• Authors: Dmitri Roussinov, Serge Sharoff
• Abstract summary: We show that classification tasks still suffer from a performance gap when the underlying distribution of topics changes. We quantify this phenomenon empirically with a large corpus and a large set of topics. We suggest and successfully test a possible remedy: after augmenting the training dataset with topically-controlled synthetic texts, the F1 score improves by up to 50% for some topics.
• Score: 0.27195102129095
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While performance of many text classification tasks has been recently improved due to Pre-trained Language Models (PLMs), in this paper we show that they still suffer from a performance gap when the underlying distribution of topics changes. For example, a genre classifier trained on \textit{political} topics often fails when tested on documents about \textit{sport} or \textit{medicine}. In this work, we quantify this phenomenon empirically with a large corpus and a large set of topics. Consequently, we verify that domain transfer remains challenging both for classic PLMs, such as BERT, and for modern large models, such as GPT-3. We also suggest and successfully test a possible remedy: after augmenting the training dataset with topically-controlled synthetic texts, the F1 score improves by up to 50\% for some topics, nearing on-topic training results, while others show little to no improvement. While our empirical results focus on genre classification, our methodology is applicable to other classification tasks such as gender, authorship, or sentiment classification. The code and data to replicate the experiments are available at https://github.com/dminus1/genre

Related papers

• Adapting Vision-Language Models to Open Classes via Test-Time Prompt Tuning [50.26965628047682]
  Adapting pre-trained models to open classes is a challenging problem in machine learning. In this paper, we consider combining the advantages of both and come up with a test-time prompt tuning approach. Our proposed method outperforms all comparison methods on average considering both base and new classes.
  arXiv  Detail & Related papers  (2024-08-29T12:34:01Z)
• Classification and Clustering of Sentence-Level Embeddings of Scientific Articles Generated by Contrastive Learning [1.104960878651584]
  Our approach consists of fine-tuning transformer language models to generate sentence-level embeddings from scientific articles. We trained our models on three datasets with contrastive learning. We show that fine-tuning sentence transformers with contrastive learning and using the generated embeddings in downstream tasks is a feasible approach to sentence classification in scientific articles.
  arXiv  Detail & Related papers  (2024-03-30T02:52:14Z)
• Language Models for Text Classification: Is In-Context Learning Enough? [54.869097980761595]
  Recent foundational language models have shown state-of-the-art performance in many NLP tasks in zero- and few-shot settings. An advantage of these models over more standard approaches is the ability to understand instructions written in natural language (prompts) This makes them suitable for addressing text classification problems for domains with limited amounts of annotated instances.
  arXiv  Detail & Related papers  (2024-03-26T12:47:39Z)
• Like a Good Nearest Neighbor: Practical Content Moderation and Text Classification [66.02091763340094]
  Like a Good Nearest Neighbor (LaGoNN) is a modification to SetFit that introduces no learnable parameters but alters input text with information from its nearest neighbor. LaGoNN is effective at flagging undesirable content and text classification, and improves the performance of SetFit.
  arXiv  Detail & Related papers  (2023-02-17T15:43:29Z)
• Many-Class Text Classification with Matching [65.74328417321738]
  We formulate textbfText textbfClassification as a textbfMatching problem between the text and the labels, and propose a simple yet effective framework named TCM. Compared with previous text classification approaches, TCM takes advantage of the fine-grained semantic information of the classification labels.
  arXiv  Detail & Related papers  (2022-05-23T15:51:19Z)
• Revisiting Self-Training for Few-Shot Learning of Language Model [61.173976954360334]
  Unlabeled data carry rich task-relevant information, they are proven useful for few-shot learning of language model. In this work, we revisit the self-training technique for language model fine-tuning and present a state-of-the-art prompt-based few-shot learner, SFLM.
  arXiv  Detail & Related papers  (2021-10-04T08:51:36Z)
• Experiments with adversarial attacks on text genres [0.0]
  Neural models based on pre-trained transformers, such as BERT or XLM-RoBERTa, demonstrate SOTA results in many NLP tasks. We show that embedding-based algorithms which can replace some of the most significant'' words with words similar to them, have the ability to influence model predictions in a significant proportion of cases.
  arXiv  Detail & Related papers  (2021-07-05T19:37:59Z)
• Pretrained Transformers for Text Ranking: BERT and Beyond [53.83210899683987]
  This survey provides an overview of text ranking with neural network architectures known as transformers. The combination of transformers and self-supervised pretraining has been responsible for a paradigm shift in natural language processing.
  arXiv  Detail & Related papers  (2020-10-13T15:20:32Z)
• Rank over Class: The Untapped Potential of Ranking in Natural Language Processing [8.637110868126546]
  We argue that many tasks which are currently addressed using classification are in fact being shoehorned into a classification mould. We propose a novel end-to-end ranking approach consisting of a Transformer network responsible for producing representations for a pair of text sequences. In an experiment on a heavily-skewed sentiment analysis dataset, converting ranking results to classification labels yields an approximately 22% improvement over state-of-the-art text classification.
  arXiv  Detail & Related papers  (2020-09-10T22:18:57Z)

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.