AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

ConZIC: Controllable Zero-Shot Image Captioning by Sampling-Based Polishing

Zequn Zeng, Hao Zhang, Ruiying Lu, Dongsheng Wang, Bo Chen, Zhengjue Wang

CVPR 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Neural image captioning · 48 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
use shuffled word-update order instead of sequential order (ablation)ConZIC on MSCOCO Karpathy test · MSCOCO Karpathy testDiv-2: 0.56 -> 0.87Improved
use shuffled word-update order instead of sequential order (ablation)ConZIC on MSCOCO Karpathy test · MSCOCO Karpathy testCLIP-S: 1.01 -> 0.99Hurt
use shuffled word-update order instead of sequential order (ablation)ConZIC on MSCOCO Karpathy test · MSCOCO Karpathy testB-4: 1.31 -> 1.29Hurt
replace left-to-right autoregressive generation with bidirectional Gibbs-BERT infillingzero-shot image captioning · MSCOCO noun infillingB-1: 0.00 -> 0.37Improved
replace left-to-right autoregressive generation with bidirectional Gibbs-BERT infillingzero-shot image captioning · MSCOCO verb infillingB-1: 0.00 -> 0.25Improved
apply a predefined POS tag sequence as a control signal (ablation)ConZIC on MSCOCO · MSCOCO POS controlAcc: 15.54 -> 83.36Improved
apply a predefined POS tag sequence as a control signal (ablation)ConZIC on MSCOCO · MSCOCO POS controlCLIP-S: 1.01 -> 0.96Hurt
add iterative polishing updates (ablation)ConZIC length-controlled captioning · MSCOCOCLIPScoreImproved
increase the caption length (ablation)ConZIC length control · MSCOCOCLIPScoreImproved
increase the corrupted-word ratio (ablation)ConZIC multiple-word infilling · MSCOCOCLIP-S: 0.83 -> 0.89 -> 0.93Improved
increase the corrupted-word ratio (ablation)ConZIC multiple-word infilling · MSCOCOB-4: 60.69 -> 26.08 -> 8.06Hurt
use ConZIC rather than ZeroCapzero-shot image captioning · SketchyCOCO captionB-1: 27.08 -> 39.61Improved

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