Research map / Vision-language and multimodal / Visual question answering
Compact Trilinear Interaction for Visual Question Answering
Tuong Do, Thanh-Toan Do, Huy Tran, Erman Tjiputra, Quang D. Tran
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 |
|---|---|---|---|
| distill the CTI trilinear teacher into the BAN2 student | BAN2 · TDIUC val (with Absurd) | Acc: 85.5 -> 87.0 | Improved |
| distill the CTI trilinear teacher into the BAN2 student | BAN2 · TDIUC val (with Absurd) | Har: 54.9 -> 65.5 | Improved |
| distill the CTI trilinear teacher into the SAN student | SAN · TDIUC val (with Absurd) | Acc: 82.3 -> 84.5 | Improved |
| distill the CTI trilinear teacher into the BAN2 student | BAN2 · TDIUC val (without Absurd) | Acc: 81.9 -> 85.0 | Improved |
| distill the CTI trilinear teacher into the SAN student | SAN · TDIUC val (without Absurd) | Acc: 79.1 -> 82.8 | Improved |
| distill the CTI teacher into BAN2 | BAN2 · TDIUC Absurd question type | Acc: 98.2 -> 93.9 | Hurt |
| distill the CTI teacher into SAN | SAN · TDIUC Absurd question type | Acc: 93.4 -> 90.6 | Hurt |
| distill the CTI teacher into BAN2 | BAN2 · TDIUC Positional Rec | Acc: 27.9 -> 40.5 | Improved |
| distill the CTI teacher into SAN | SAN · TDIUC Sport Rec | Acc: 95.5 -> 95.5 | No clear effect |
| distill the CTI teacher into BAN2 | BAN2 · VQA-2.0 test-dev | Acc: 66.5 -> 67.4 | Improved |
| distill the CTI teacher into SAN | SAN · VQA-2.0 test-dev | Acc: 63.0 -> 63.4 | Improved |
| replace BAN2 with the CTI trilinear model | BAN2 · Visual7W val | Acc-MC: 65.7 -> 67.0 | Improved |
Closest papers by meaning
- MUTAN: Multimodal Tucker Fusion for Visual Question Answering (ICCV 2017 · similarity 0.91)
- An Analysis of Visual Question Answering Algorithms (ICCV 2017 · similarity 0.90)
- Answer-Type Prediction for Visual Question Answering (CVPR 2016 · similarity 0.89)
- Multi-Modality Latent Interaction Network for Visual Question Answering (ICCV 2019 · similarity 0.89)
- Bridging the Gap between 2D and 3D Visual Question Answering: A Fusion Approach for 3D VQA (AAAI 2024 · similarity 0.89)
- When Open-Vocabulary Visual Question Answering Meets Causal Adapter: Benchmark and Approach (AAAI 2025 · similarity 0.89)
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.
