AI Paper Atlas

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

Can Vision-Language Models Answer Face to Face Questions in the Real-World?

Reza Pourreza, Rishit Dagli, Apratim Bhattacharyya, Sunny Panchal, Guillaume Berger, Roland Memisevic

ICLR 2026 · Vision-language and multimodal › Visual question answering · approach: Vision-language model VQA evaluation

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 audio input to VideoLLaMA2.1-7B-AV (pretrained, zero-shot) (ablation)VideoLLaMA2.1-7B-AV · QIVDcorrectnessHurt
fine-tune VideoLLaMA2.1-7B-AV on QIVD with audio and video (5-fold CV) (ablation)VideoLLaMA2.1-7B-AV · QIVDcorrectness: action counting +16.96%, subjective +23.26%, audio-visual +17.39%Improved
fine-tune on QIVD, object attributes and scene understanding (ablation)VideoLLaMA2.1-7B-AV · QIVD object attributes / scene understandingcorrectness: +1.24% / +2.63%No clear effect
fine-tuned model evaluated on video-only input (ablation)fine-tuned VideoLLaMA2.1-7B-AV · QIVD video-onlycorrectnessHurt
use fine-tuned Stream-Qwen-Omni to detect when-to-answer instead of Whisper-StreamingQwen2.5-Omni-7B · QIVDwhen-to-answer MAE (s): 0.83 -> 0.52Improved
provide accurate when-to-answer timestamp (ablation)Qwen2.5-Omni-7B · QIVDcorrectnessImproved
use ground-truth instead of ASR questions and timestamps (ablation)baseline LMMs · QIVDcorrectness: Qwen3-VL-8B 53.72 -> 60.07Improved
restrict frames to +-0.5 s around question instead of full video (ablation)Qwen2.5-VL-7B · QIVDcorrectness: 60.00 -> 38.28Hurt
widen frame window to +-3 s around question (ablation)Qwen2.5-VL-7B · QIVDcorrectness: 60.00 -> 61.20No clear effect
move from static to temporal question categoriesall evaluated LMMs · QIVDcorrectness: GPT-4o 65.24 -> 46.06Hurt
scale model to VideoLLaMA2-72B from 7B (ablation)VideoLLaMA2 · QIVDcorrectness: 50.07 -> 50.83Improved

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