Examining Cooperation in Visual Dialog Models

December 04, 2017 · Entered Twilight · 🏛 arXiv.org

"Last commit was 8.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .gitignore, IO.py, LICENSE, README.md, evaluate.py, models.py, train.py

Authors Mircea Mironenco, Dana Kianfar, Ke Tran, Evangelos Kanoulas, Efstratios Gavves arXiv ID 1712.01329 Category cs.CV: Computer Vision Cross-listed cs.AI, cs.CL, cs.IR, cs.LG Citations 4 Venue arXiv.org Repository https://github.com/danakianfar/Examining-Cooperation-in-VDM/ ⭐ 1 Last Checked 1 month ago

Abstract

In this work we propose a blackbox intervention method for visual dialog models, with the aim of assessing the contribution of individual linguistic or visual components. Concretely, we conduct structured or randomized interventions that aim to impair an individual component of the model, and observe changes in task performance. We reproduce a state-of-the-art visual dialog model and demonstrate that our methodology yields surprising insights, namely that both dialog and image information have minimal contributions to task performance. The intervention method presented here can be applied as a sanity check for the strength and robustness of each component in visual dialog systems.