Evaluating Large Language Models Using Contrast Sets: An Experimental Approach

• URL: http://arxiv.org/abs/2404.01569v2
• Date: Wed, 02 Oct 2024 12:31:11 GMT
• Title: Evaluating Large Language Models Using Contrast Sets: An Experimental Approach
• Authors: Manish Sanwal,
• Abstract summary: We introduce an innovative technique for generating a contrast set for the Stanford Natural Language Inference dataset. Our strategy involves the automated substitution of verbs, adverbs, and adjectives with their synonyms to preserve the original meaning of sentences. This method aims to assess whether a model's performance is based on genuine language comprehension or simply on pattern recognition.
• Score: 0.0
• License:
• Abstract: In the domain of Natural Language Inference (NLI), especially in tasks involving the classification of multiple input texts, the Cross-Entropy Loss metric is widely employed as a standard for error measurement. However, this metric falls short in effectively evaluating a model's capacity to understand language entailments. In this study, we introduce an innovative technique for generating a contrast set for the Stanford Natural Language Inference (SNLI) dataset. Our strategy involves the automated substitution of verbs, adverbs, and adjectives with their synonyms to preserve the original meaning of sentences. This method aims to assess whether a model's performance is based on genuine language comprehension or simply on pattern recognition. We conducted our analysis using the ELECTRA-small model. The model achieved an accuracy of 89.9% on the conventional SNLI dataset but showed a reduced accuracy of 72.5% on our contrast set, indicating a substantial 17% decline. This outcome led us to conduct a detailed examination of the model's learning behaviors. Following this, we improved the model's resilience by fine-tuning it with a contrast-enhanced training dataset specifically designed for SNLI, which increased its accuracy to 85.5% on the contrast sets. Our findings highlight the importance of incorporating diverse linguistic expressions into datasets for NLI tasks. We hope that our research will encourage the creation of more inclusive datasets, thereby contributing to the development of NLI models that are both more sophisticated and effective.

Related papers

• How Hard is this Test Set? NLI Characterization by Exploiting Training Dynamics [49.9329723199239]
  We propose a method for the automated creation of a challenging test set without relying on the manual construction of artificial and unrealistic examples. We categorize the test set of popular NLI datasets into three difficulty levels by leveraging methods that exploit training dynamics. When our characterization method is applied to the training set, models trained with only a fraction of the data achieve comparable performance to those trained on the full dataset.
  arXiv  Detail & Related papers  (2024-10-04T13:39:21Z)
• ECRC: Emotion-Causality Recognition in Korean Conversation for GCN [0.0]
  We propose the emotion-causality recognition in conversation (ECRC) model, which is based on a novel graph structure. In this study, we overcome the limitations of previous embeddings by utilizing both word- and sentence-level embeddings. This model uniquely integrates the bidirectional long short-term memory (Bi-LSTM) and graph neural network (GCN) models for Korean conversation analysis.
  arXiv  Detail & Related papers  (2024-03-16T02:07:31Z)
• Dissecting vocabulary biases datasets through statistical testing and automated data augmentation for artifact mitigation in Natural Language Inference [3.154631846975021]
  We focus on investigating dataset artifacts and developing strategies to address these issues. We propose several automatic data augmentation strategies spanning character to word levels. Experiments demonstrate that the proposed approaches effectively enhance model accuracy and reduce biases by up to 0.66% and 1.14%, respectively.
  arXiv  Detail & Related papers  (2023-12-14T08:46:26Z)
• Influence Scores at Scale for Efficient Language Data Sampling [3.072340427031969]
  "influence scores" are used to identify important subsets of data. In this paper, we explore the applicability of influence scores in language classification tasks.
  arXiv  Detail & Related papers  (2023-11-27T20:19:22Z)
• Fine-tuning Language Models for Factuality [96.5203774943198]
  Large pre-trained language models (LLMs) have led to their widespread use, sometimes even as a replacement for traditional search engines. Yet language models are prone to making convincing but factually inaccurate claims, often referred to as 'hallucinations' In this work, we fine-tune language models to be more factual, without human labeling.
  arXiv  Detail & Related papers  (2023-11-14T18:59:15Z)
• A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss [17.213602354715956]
  We propose a BERT-based model with feature projection and length-balanced loss for readability assessment. Our model achieves state-of-the-art performances on two English benchmark datasets and one dataset of Chinese textbooks.
  arXiv  Detail & Related papers  (2022-10-19T05:33:27Z)
• An Empirical Investigation of Commonsense Self-Supervision with Knowledge Graphs [67.23285413610243]
  Self-supervision based on the information extracted from large knowledge graphs has been shown to improve the generalization of language models. We study the effect of knowledge sampling strategies and sizes that can be used to generate synthetic data for adapting language models.
  arXiv  Detail & Related papers  (2022-05-21T19:49:04Z)
• 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)
• Understanding and Improving Lexical Choice in Non-Autoregressive Translation [98.11249019844281]
  We propose to expose the raw data to NAT models to restore the useful information of low-frequency words. Our approach pushes the SOTA NAT performance on the WMT14 English-German and WMT16 Romanian-English datasets up to 27.8 and 33.8 BLEU points, respectively.
  arXiv  Detail & Related papers  (2020-12-29T03:18:50Z)
• Comparison of Interactive Knowledge Base Spelling Correction Models for Low-Resource Languages [81.90356787324481]
  Spelling normalization for low resource languages is a challenging task because the patterns are hard to predict. This work shows a comparison of a neural model and character language models with varying amounts on target language data. Our usage scenario is interactive correction with nearly zero amounts of training examples, improving models as more data is collected.
  arXiv  Detail & Related papers  (2020-10-20T17:31:07Z)
• Comparative Study of Language Models on Cross-Domain Data with Model Agnostic Explainability [0.0]
  The study compares the state-of-the-art language models - BERT, ELECTRA and its derivatives which include RoBERTa, ALBERT and DistilBERT. The experimental results establish new state-of-the-art for 2013 rating classification task and Financial Phrasebank sentiment detection task with 69% accuracy and 88.2% accuracy respectively.
  arXiv  Detail & Related papers  (2020-09-09T04:31:44Z)

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.