AI Paper Atlas

Research map / Vision-language and multimodal / Visual question answering

Guessing State Tracking for Visual Dialogue

Wei Pang, Xiaojie Wang

ECCV 2020 · Vision-language and multimodal › Visual question answering · approach: Visual reasoning diagnostics · 11 citations (OpenAlex)

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 changedWhereMeasuredEffect
add early supervision and plain supervision (ES&PS) (ablation)GST guesser (SL) · GuessWhat?! NewObject, Greedyguessing success rate (%): 42.58 -> 54.10Improved
add early supervision and plain supervision (ES&PS) (ablation)GST guesser (SL) · GuessWhat?! NewGame, Greedyguessing success rate (%): 39.48 -> 50.97Improved
add incremental supervision (IS) (ablation)GST guesser (SL) · GuessWhat?! NewObject, Greedyguessing success rate (%): 53.49 -> 54.10Improved
add incremental supervision (IS) (ablation)GST guesser (SL) · GuessWhat?! NewGame, Greedyguessing success rate (%): 50.33 -> 50.97Improved
use symmetric concatenation [h_qa; h_qa*O; O] instead of [h_qa; O] (ablation)GST guesser (SL) · GuessWhat?! testerror rate (%): 37.8 -> 34.4Improved
use symmetric concatenation instead of only element-wise product (ablation)GST guesser (SL) · GuessWhat?! testerror rate (%): 36.7 -> 34.4Improved
raise c in the log term from 1.1 to 1.5 or 2.0 (ablation)GST guesser (SL) · GuessWhat?! testerror rate (%): 34.3 -> 34.8No clear effect
replace single-step guesser with guessing state tracking (GST), VDST QGen, 8 questions, RLguesser (previous) · GuessWhat?! NewObject, Greedyguessing success rate (%): 70.55 -> 83.32Improved
replace single-step guesser with GST, VDST QGen, 5 questions, SLguesser (previous) · GuessWhat?! NewObject, Greedyguessing success rate (%): 49.49 -> 53.35Improved
replace single-step guesser with GST, qgen, 5 questions, RLguesser (previous) · GuessWhat?! NewObject, Greedyguessing success rate (%): 60.3 -> 67.06Improved
increase maximum questions from 5 to 8 for GST, RLGST + VDST · GuessWhat?! NewObject, Greedyguessing success rate (%): 77.30 -> 83.32Improved
GST trained by RL instead of SL (8 questions)GST · GuessWhat?! testerror rate (%): 34.3 -> 18.4Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy