Results 21 to 30 of about 271,354 (322)
Few-Shot Self-Rationalization with Natural Language Prompts [PDF]
NAACL-HLT, 2021 Self-rationalization models that predict task labels and generate free-text elaborations for their predictions could enable more intuitive interaction with NLP systems.
Ana Marasović +3 more
semanticscholar +1 more source
Can Rationalization Improve Robustness? [PDF]
North American Chapter of the Association for Computational Linguistics, 2022 A growing line of work has investigated the development of neural NLP models that can produce rationales–subsets of input that can explain their model predictions.
Howard Chen +3 more
semanticscholar +1 more source
FR: Folded Rationalization with a Unified Encoder [PDF]
Neural Information Processing Systems, 2022 Conventional works generally employ a two-phase model in which a generator selects the most important pieces, followed by a predictor that makes predictions based on the selected pieces.
Wei Liu +5 more
semanticscholar +1 more source
International Journal of Environmental Research and Public Health, 2022 Carbon dioxide mainly comes from industrial economic activities. Industrial structure optimization is an effective way to reduce carbon dioxide emissions.
Runde Gu +4 more
semanticscholar +1 more source
Does Self-Rationalization Improve Robustness to Spurious Correlations? [PDF]
Conference on Empirical Methods in Natural Language Processing, 2022 Rationalization is fundamental to human reasoning and learning. NLP models trained to produce rationales along with predictions, called self-rationalization models, have been investigated for their interpretability and utility to end-users.
Alexis Ross +2 more
semanticscholar +1 more source
Distribution Matching for Rationalization [PDF]
AAAI Conference on Artificial Intelligence, 2021 The task of rationalization aims to extract pieces of input text as rationales to justify neural network predictions on text classification tasks. By definition, rationales represent key text pieces used for prediction and thus should have similar ...
Yongfeng Huang +3 more
semanticscholar +1 more source
Interventional Rationalization
Conference on Empirical Methods in Natural Language Processing, 2023 Selective rationalizations improve the explain-ability of neural networks by selecting a sub-sequence of the input (i.e., rationales) to explain the prediction results.
Linan Yue +5 more
semanticscholar +1 more source
SPECTRA: Sparse Structured Text Rationalization [PDF]
Conference on Empirical Methods in Natural Language Processing, 2021 Selective rationalization aims to produce decisions along with rationales (e.g., text highlights or word alignments between two sentences). Commonly, rationales are modeled as stochastic binary masks, requiring sampling-based gradient estimators, which ...
Nuno M. Guerreiro, André F. T. Martins
semanticscholar +1 more source
Rationalization through Concepts [PDF]
Findings, 2021 Automated predictions require explanations to be interpretable by humans. One type of explanation is a rationale, i.e., a selection of input features such as relevant text snippets from which the model computes the outcome.
Diego Antognini, B. Faltings
semanticscholar +1 more source
International Journal of Environmental Research and Public Health, 2022 Whistleblowers who expose wrongdoing often face several concerns, pressures, and threats of retaliation before reaching a final decision. Specifically, this study examines the effects of perceived seriousness of wrongdoing (PSW) and perceived threat of ...
J. Khan +6 more
semanticscholar +1 more source