Research map / Label-efficient and robust learning / Recognition and long-tail
FLEX: Faithful Linguistic Explanations for Neural Net Based Model Decisions
Sandareka Wickramanayake, Wynne Hsu, Mong Li Lee
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 relevance vector and relevance loss to FLEX (ablation) | FLEX · CUB | DREL: 16.48 -> 17.85 | Improved |
| add relevance vector and relevance loss to FLEX (ablation) | FLEX · MPII | DREL: 14.13 -> 16.11 | Improved |
| add weighted CNN features to baseline captioner (ablation) | Baseline · CUB | DREL: 13.67 -> 16.48 | Improved |
| add weighted CNN features to baseline captioner (ablation) | Baseline · MPII | DREL: 13.48 -> 14.13 | Improved |
| vary relevance loss weight lambda from 0 to 0.1 (ablation) | FLEX · CUB | DREL: 16.48 -> 17.85 | Improved |
| use FLEX instead of GVE on CUB | explanation generator · CUB | DREL: 15.67 -> 17.85 | Improved |
| use FLEX instead of MME on MPII | explanation generator · MPII | BLEU-4: 19.88 -> 19.11 | Hurt |
| use FLEX instead of GVE on CUB | explanation generator · CUB | BLEU-4: 28.43 -> 30.16 | Improved |
| use FLEX instead of GVE/MME for unseen classes | explanation generator · CUB unseen classes | DREL: 32.35 -> 36.72 | Improved |
Closest papers by meaning
- FaCT: Faithful Concept Traces for Explaining Neural Network Decisions (NeurIPS 2025 · similarity 0.87)
- Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations (ICML 2025 · similarity 0.87)
- Explainable Neural Networks with Guarantee: A Sparse Estimation Approach (AAAI 2025 · similarity 0.86)
- Keep the Faith: Faithful Explanations in Convolutional Neural Networks for Case-Based Reasoning (AAAI 2024 · similarity 0.86)
- Explain Yourself, Briefly! Self-Explaining Neural Networks with Concise Sufficient Reasons (ICLR 2025 · similarity 0.85)
- Towards Modeling Uncertainties of Self-Explaining Neural Networks via Conformal Prediction (AAAI 2024 · similarity 0.85)
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.
