AI Paper Atlas

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

Transferable Decoding with Visual Entities for Zero-Shot Image Captioning

Junjie Fei, Teng Wang, Jinrui Zhang, Zhenyu He, Chengjie Wang, Feng Zheng

ICCV 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 88 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 entity-aware hard prompts (with masking) to CapDec-style soft prompts (ablation)CapDec · COCO => NoCaps val out-of-domainCIDEr: 28.7 -> 65.0Improved
add entity masking (40%) to entity hard prompts (ablation)ViECap · COCO testCIDEr: 53.2 -> 92.9Improved
increase entity masking ratio beyond 40% (ablation)ViECap · NoCaps val overallCIDEr: 66.2 -> 65.5 -> 65.3 (60%, 80%)Mixed
combine soft and hard prompts instead of entity-only or soft-only (ablation)ViECap · NoCaps val overallCIDEr: 45.9 (soft) / 44.4 (entity) -> 66.2Improved
use CLIP prompt ensemble for entity retrieval (ablation)ViECap · NoCaps val overallCIDEr: 61.0 -> 66.2Improved
scale up frozen language model from GPT-2 Base to OPT-2.7B (ablation)ViECap · NoCaps val overallCIDEr: 62.0 -> 72.1Improved
increase soft prompt length from 10 to 20, 30, 40 (ablation)ViECap · NoCaps val overallCIDEr: 66.2 vs 65.6 / 66.0 / 65.9No clear effect
change hard prompt template wording (ablation)ViECap · NoCaps val overallCIDEr: 66.2 vs 65.2 / 65.3 / 64.9No clear effect
use large noisy Visual Genome vocabulary instead of cleaned VGOI vocabulary (ablation)ViECap · NoCaps val overallCIDEr: 66.2 -> 50.0Hurt
add entity-aware hard prompts under in-domain captioningCapDec · Flickr30k testCIDEr: 39.1 -> 47.9Improved
train with 0.1% of COCO dataViECap vs CapDec · NoCaps val overallCIDEr: 10.4 -> 30.2Improved
train on 10% COCO dataViECap vs CapDec · COCO testCIDEr: 83.6 vs 83.4No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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