AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

BEAF: Observing BEfore-AFter Changes to Evaluate Hallucination in Vision-language Models

Moon Ye-Bin, Nam Hyeon-Woo, Wonseok Choi, Tae-Hyun Oh

ECCV 2024 · Vision-language and multimodal › CLIP and zero-shot · approach: Adapting VLMs for zero-shot classification · 10 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
evaluate with change-aware TU instead of traditional accuracyVLM hallucination evaluation · BEAFTU vs accuracy: accuracy 74.9 vs TU 24.3 (LLaVA-v1.5 13B)Mixed
location-aware training (Shikra) versus other 7B VLMsShikra · BEAFTU: 52.7 vs 33.5 (InstructBLIP 7B)Improved
larger model size 7B to 13B (LLaVA-v1.5)LLaVA-v1.5 · BEAFTU: 32.6 -> 24.3Hurt
higher VQAv2/GQA performance versus hallucination resistanceVLMs 7B · VQAv2 and BEAFTU: mPLUG-Owl2 VQAv2 79.4 but TU 24.6Hurt
remove object from image via inpaintingBEAF images · BEAFCLIP Score: 53.7 -> 47.9Improved

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