Self-Rationalization in the Wild: A Large Scale Out-of-Distribution Evaluation on NLI-related tasks

• URL: http://arxiv.org/abs/2502.04797v1
• Date: Fri, 07 Feb 2025 10:01:32 GMT
• Title: Self-Rationalization in the Wild: A Large Scale Out-of-Distribution Evaluation on NLI-related tasks
• Authors: Jing Yang, Max Glockner, Anderson Rocha, Iryna Gurevych,
• Abstract summary: Free-text explanations are expressive and easy to understand, but many datasets lack annotated explanation data. We fine-tune T5-Large and OLMo-7B models and assess the impact of fine-tuning data quality, the number of fine-tuning samples, and few-shot selection methods. The models are evaluated on 19 diverse OOD datasets across three tasks: natural language inference (NLI), fact-checking, and hallucination detection in abstractive summarization.
• Score: 59.47851630504264
• License:
• Abstract: Free-text explanations are expressive and easy to understand, but many datasets lack annotated explanation data, making it challenging to train models for explainable predictions. To address this, we investigate how to use existing explanation datasets for self-rationalization and evaluate models' out-of-distribution (OOD) performance. We fine-tune T5-Large and OLMo-7B models and assess the impact of fine-tuning data quality, the number of fine-tuning samples, and few-shot selection methods. The models are evaluated on 19 diverse OOD datasets across three tasks: natural language inference (NLI), fact-checking, and hallucination detection in abstractive summarization. For the generated explanation evaluation, we conduct a human study on 13 selected models and study its correlation with the Acceptability score (T5-11B) and three other LLM-based reference-free metrics. Human evaluation shows that the Acceptability score correlates most strongly with human judgments, demonstrating its effectiveness in evaluating free-text explanations. Our findings reveal: 1) few annotated examples effectively adapt models for OOD explanation generation; 2) compared to sample selection strategies, fine-tuning data source has a larger impact on OOD performance; and 3) models with higher label prediction accuracy tend to produce better explanations, as reflected by higher Acceptability scores.

Related papers

• A Comprehensive Evaluation and Analysis Study for Chinese Spelling Check [53.152011258252315]
  We show that using phonetic and graphic information reasonably is effective for Chinese Spelling Check. Models are sensitive to the error distribution of the test set, which reflects the shortcomings of models. The commonly used benchmark, SIGHAN, can not reliably evaluate models' performance.
  arXiv  Detail & Related papers  (2023-07-25T17:02:38Z)
• Think Twice: Measuring the Efficiency of Eliminating Prediction Shortcuts of Question Answering Models [3.9052860539161918]
  We propose a simple method for measuring a scale of models' reliance on any identified spurious feature. We assess the robustness towards a large set of known and newly found prediction biases for various pre-trained models and debiasing methods in Question Answering (QA) We find that while existing debiasing methods can mitigate reliance on a chosen spurious feature, the OOD performance gains of these methods can not be explained by mitigated reliance on biased features.
  arXiv  Detail & Related papers  (2023-05-11T14:35:00Z)
• Are Sample-Efficient NLP Models More Robust? [90.54786862811183]
  We investigate the relationship between sample efficiency (amount of data needed to reach a given ID accuracy) and robustness (how models fare on OOD evaluation) We find that higher sample efficiency is only correlated with better average OOD robustness on some modeling interventions and tasks, but not others. These results suggest that general-purpose methods for improving sample efficiency are unlikely to yield universal OOD robustness improvements, since such improvements are highly dataset- and task-dependent.
  arXiv  Detail & Related papers  (2022-10-12T17:54:59Z)
• 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)
• Understanding and Testing Generalization of Deep Networks on Out-of-Distribution Data [30.471871571256198]
  Deep network models perform excellently on In-Distribution data, but can significantly fail on Out-Of-Distribution data. This study is devoted to analyzing the problem of experimental ID test and designing OOD test paradigm.
  arXiv  Detail & Related papers  (2021-11-17T15:29:07Z)
• How can I choose an explainer? An Application-grounded Evaluation of Post-hoc Explanations [2.7708222692419735]
  Explanations are seldom evaluated based on their true practical impact on decision-making tasks. This study proposes XAI Test, an application-grounded evaluation methodology tailored to isolate the impact of providing the end-user with different levels of information. Using strong statistical analysis, we show that, in general, popular explainers have a worse impact than desired.
  arXiv  Detail & Related papers  (2021-01-21T18:15:13Z)
• Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language? [86.60613602337246]
  We introduce a leakage-adjusted simulatability (LAS) metric for evaluating NL explanations. LAS measures how well explanations help an observer predict a model's output, while controlling for how explanations can directly leak the output. We frame explanation generation as a multi-agent game and optimize explanations for simulatability while penalizing label leakage.
  arXiv  Detail & Related papers  (2020-10-08T16:59:07Z)
• What Can We Learn from Collective Human Opinions on Natural Language Inference Data? [88.90490998032429]
  ChaosNLI is a dataset with a total of 464,500 annotations to study Collective HumAn OpinionS. This dataset is created by collecting 100 annotations per example for 3,113 examples in SNLI and MNLI and 1,532 examples in Abductive-NLI.
  arXiv  Detail & Related papers  (2020-10-07T17:26:06Z)
• Learning by Semantic Similarity Makes Abstractive Summarization Better [13.324006587838522]
  We compare the generated summaries from recent LM, BART, and the reference summaries from a benchmark dataset, CNN/DM. Interestingly, model-generated summaries receive higher scores relative to reference summaries.
  arXiv  Detail & Related papers  (2020-02-18T17:59:02Z)

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.