Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
MARTA: Leveraging Human Rationales for Explainable Text Classification
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add human rationales and model worker reliability during attention-based training (ablation) | Attention-based text classification · Wiki-Tech | Accuracy: 23% | Helped |
| Add human rationales and model worker reliability during attention-based training (ablation) | Attention-based text classification · Wiki-Tech | F1-score: 28% | Helped |
| Add human rationales and model worker reliability during attention-based training (ablation) | Attention-based text classification · Amazon | Accuracy: 12.5% | Helped |
| Add human rationales and model worker reliability during attention-based training (ablation) | Attention-based text classification · Amazon | F1-score: 12.5% | Helped |
| Use MARTA rather than ALBERT | Text classification · Wiki-Tech and Amazon | Accuracy: 0.97% | Helped |
| Use MARTA rather than ALBERT | Text classification · Wiki-Tech and Amazon | F1-score: 5.76% | Helped |
| Use MARTA rather than LSTM-diversity | Text classification · Wiki-Tech and Amazon | Accuracy: 18.68% | Helped |
| Use MARTA rather than LSTM-diversity | Text classification · Wiki-Tech and Amazon | F1-score: 17.61% | Helped |
| Train with human rationales and reliability-weighted attention instead of the attention-based model alone (ablation) | Attention-based text classification · Wiki-Tech | Rationale recall: 45.1% | Helped |
| Train with human rationales and reliability-weighted attention instead of the attention-based model alone (ablation) | Attention-based text classification · Amazon | Rationale recall: 22.6% | Helped |
Closest papers by meaning
- Automated Assessment of Fidelity and Interpretability: An Evaluation Framework for Large Language Models’ Explanations (Student Abstract) (AAAI 2024 · similarity 0.85)
- Learning Robust Rationales for Model Explainability: A Guidance-Based Approach (AAAI 2024 · similarity 0.84)
- A General Search-Based Framework for Generating Textual Counterfactual Explanations (AAAI 2024 · similarity 0.84)
- NOMATTERXAI: Generating “No Matter What” Alterfactual Examples for Explaining Black-Box Text Classification Models (AAAI 2025 · similarity 0.83)
- Towards Unifying Evaluation of Counterfactual Explanations: Leveraging Large Language Models for Human-Centric Assessments (AAAI 2025 · similarity 0.83)
- De-biased Attention Supervision for Text Classification with Causality (AAAI 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.