A Unified Understanding of Deep NLP Models for Text Classification

Li, Zhen; Wang, Xiting; Yang, Weikai; Wu, Jing; Zhang, Zhengyan; Liu, Zhiyuan; Sun, Maosong; Zhang, Hui; Liu, Shixia

doi:10.1109/tvcg.2022.3184186

Cited by 30 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sprinner et al [71] designed a framework for explainable artificial intelligence (XAI) and operationalized it as explAIner, plugged into TensorBoard [1], the most widely used platform for model analysis and visualization. As for the XAI system for NLP, Li et al [45] provided a unified interpretive method for interpreting NLP models for text classification. Attempts have also been made in the broaderer application scenarios of AI, such as healthcare [9] and autonomous driving [28], [83].…”

Section: Visual Explanation For Machine Learningmentioning

confidence: 99%

“…Several visual analytic systems for explaining machine learning have been developed. For example, CNNVis [47] tries to help experts analyze CNNs by converting structures into directed acyclic graphs combined with various algorithms and Li et al [45] proposes a unified structure to interpret deep NLP models for text classification. However, their method focuses on explaining a single model and is limited in exploring the internals of a multi-module model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Visual Explanation for Open-Domain Question Answering With BERT

Zekai

Sun

Zhao

et al. 2024

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

show abstract

Section: Visual Explanation For Machine Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Visual Explanation for Open-Domain Question Answering With BERT

Zekai

Sun

Zhao

et al. 2024

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

show abstract

“…For example, visualisation methods highlight the parts of a text (e.g., a student answer) that most contributed to the final output (e.g., a mark). Software is being developed to allow researchers to see what the AI considered the most important words in a text to be (Li et al, 2022; Tenney et al, 2020). These new tools have enabled researchers to probe black‐box systems by comparing machine and human explanations.…”

Section: Barriers For Using Artificial Intelligence and Machine Learn...mentioning

confidence: 99%

The future of standardised assessment: Validity and trust in algorithms for assessment and scoring

Aloisi¹

2023

Euro J of Education

View full text Add to dashboard Cite

This article considers the challenges of using artificial intelligence (AI) and machine learning (ML) to assist high‐stakes standardised assessment. It focuses on the detrimental effect that even state‐of‐the‐art AI and ML systems could have on the validity of national exams of secondary education, and how lower validity would negatively affect trust in the system. To reach this conclusion, three unresolved issues in AI (unreliability, low explainability and bias) are addressed, to show how each of them would compromise the interpretations and uses of exam results (i.e., exam validity). Furthermore, the article relates validity to trust, and specifically to the ABI+ model of trust. Evidence gathered as part of exam validation supports each of the four trust‐enabling components of the ABI+ model (ability, benevolence, integrity and predictability). It is argued, therefore, that the three AI barriers to exam validity limit the extent to which an AI‐assisted exam system could be trusted. The article suggests that addressing the issues of AI unreliability, low explainability and bias should be sufficient to put AI‐assisted exams on par with traditional ones, but might not go as far as fully reassure the public. To achieve this, it is argued that changes to the quality assurance mechanisms of the exam system will be required. This may involve, for example, integrating principled AI frameworks in assessment policy and regulation.

show abstract

“…While many visualizations have been built to help practitioners evaluate models over time, one area of recent work has focused on designing and developing analytic tools for ML error discovery [e.g. 25,63,66,82,132,136]. For example, FairVis [19] uses visualizations to help ML developers discover model bias by investigating known subgroups and exploring similar groups in the tabular data.…”

Section: Visualizationmentioning

confidence: 99%

Angler: Helping Machine Translation Practitioners Prioritize Model Improvements

Robertson

Wang

Moritz

et al. 2023

Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

Angler enables ML developers to easily explore and curate challenge sets for machine translation. (A) The Table View lists all challenge sets, allowing users to compare them by metrics such as sample count, model performance, and familiarity score. After selecting a set, (B) the Detail View allows users to further explore samples in this set across various dimensions. (B1) The Timeline enables users to query data samples by time. (B2) The Spotlight presents visualizations with linking and brushing to help users characterize the set from different angles. (B3) The Sentence List shows all selected data samples and allows users to further fine-tune before exporting this challenge set for downstream tasks.

show abstract

A Unified Understanding of Deep NLP Models for Text Classification

Cited by 30 publications

References 37 publications

Visual Explanation for Open-Domain Question Answering With BERT

Visual Explanation for Open-Domain Question Answering With BERT

The future of standardised assessment: Validity and trust in algorithms for assessment and scoring

Angler: Helping Machine Translation Practitioners Prioritize Model Improvements

Contact Info

Product

Resources

About