AI Paper Atlas

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

Affective Visual Dialog: A Large-Scale Benchmark for Emotional Reasoning Based on Visually Grounded Conversations

Kilichbek Haydarov, Xiaoqian Shen, Avinash Madasu, Mahmoud Salem, Li-Jia Li, Gamaleldin F Elsayed, Mohamed Elhoseiny

ECCV 2024 · Vision-language and multimodal › Visual question answering · approach: Visual reasoning diagnostics · 7 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 dialog D to the input of explanation generation models (ablation)T5-Large · AffectVisDial Questioner explanationBLEU: 0.05 -> 0.26Improved
add dialog D to the input (ablation)BART-Large · AffectVisDial Questioner explanationEmo-F1: 2.10 -> 51.62Improved
add dialog D to input of zero-shot GPT-4 (ablation)GPT-4 zero-shot · AffectVisDialEmo-F1: 24.28 -> 29.79Improved
add image description I to text-only models (ablation)T5-Large / BART (no dialog) · AffectVisDial Questioner explanationBLEU / Emo-F1: Emo-F1 15.20 -> 38.11 (T5, no dialog)Improved
add image description I when dialog is already present (ablation)BART-Large · AffectVisDial Questioner explanationEmo-F1: 51.62 -> 50.43No clear effect
remove emotion labels E from the input (ablation)BART-Large · AffectVisDial Questioner explanationEmo-F1: 51.62 -> 47.49Hurt
fine-tune MiniGPT-4-v2 on AffectVisDial instead of zero-shotMiniGPT-4-v2 · AffectVisDial explanationEmo-F1: 25.28 -> 46.62Improved
use discriminative instead of generative models for dialog QALTMI · AffectVisDial dialog QAR@1: 0.18 -> 0.25Improved
fine-tune on AffectVisDial for G-Eval groundedness and relevanceLLaMa2-7b · G-EvalRelevance: 2.02 -> 2.34Improved
fine-tune LLaMa2-7b on AffectVisDial for coherence and fluencyLLaMa2-7b · G-EvalFluency: 3.10 -> 2.83Hurt

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