Research map / Vision-language and multimodal / Visual question answering
Attentive User-Engaged Adversarial Neural Network for Community Question Answering
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 user context (historical answers) (ablation) | AUANN · SemEval-2016 Task 3 test | Accuracy: 77.73 -> 80.49 | Helped |
| add user context (historical answers) (ablation) | AUANN · SemEval-2017 Task 3 test | F1: 77.98 -> 79.81 | Helped |
| add adversarial discriminator for user context denoising (ablation) | AUANN · SemEval-2016 Task 3 test | Accuracy: 76.09 -> 80.49 | Helped |
| add adversarial discriminator for user context denoising (ablation) | AUANN · SemEval-2017 Task 3 test | F1: 78.14 -> 79.81 | Helped |
| use user context without discriminator (vs no user context) (ablation) | basic model · SemEval-2016 Task 3 test | Accuracy: 77.73 -> 76.09 | Hurt |
| use similarity-based selection instead of random selection of user answers (ablation) | AUANN · SemEval-2016 Task 3 test | Accuracy: 76.54 -> 80.49 | Helped |
| add semantic self-attention (ablation) | AUANN · SemEval-2016 Task 3 test | Accuracy: 78.17 -> 80.49 | Helped |
| add co-attention (ablation) | AUANN · SemEval-2017 Task 3 test | F1: 77.20 -> 79.81 | Helped |
| add answerer/anonymous metadata features (ablation) | AUANN · SemEval-2016 Task 3 test | Accuracy: 78.20 -> 80.49 | Helped |
Closest papers by meaning
- Adversarial Training for Community Question Answer Selection Based on Multi-Scale Matching (AAAI 2019 · similarity 0.89)
- Attentive Interactive Neural Networks for Answer Selection in Community Question Answering (AAAI 2017 · similarity 0.88)
- Hybrid Attentive Answer Selection in CQA With Deep Users Modelling (AAAI 2018 · similarity 0.88)
- Human-Adversarial Visual Question Answering (NeurIPS 2021 · similarity 0.86)
- VQAttack: Transferable Adversarial Attacks on Visual Question Answering via Pre-trained Models (AAAI 2024 · similarity 0.83)
- NaturalBench: Evaluating Vision-Language Models on Natural Adversarial Samples (NeurIPS 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.