AI Paper Atlas

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

Instruction-Guided Scene Text Recognition

Yongkun Du, Zhineng Chen, Yuchen Su, Caiyan Jia, Yu–Gang Jiang

TPAMI 2025 · Vision-language and multimodal › Human and embodied modeling · approach: Language-grounded scene text and structure · 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
add frequency, position and sub-string attribute questions to recognition instructions (ablation)IGTR-AR · Union14M-L test (Union14M-L-trained)accuracy: 81.27 -> 84.02Improved
add all attribute questions (full instructions) (ablation)IGTR-PR · Union14M-L test (Union14M-L-trained)accuracy: +1.94%Improved
add condition modes cond-1 to cond-4 (ablation)IGTR-PR · Union14M-L test (Union14M-L-trained)accuracy: 79.02 -> 82.23Improved
increase number of partitions K from 2 to 6 (ablation)IGTR-AR · Union14M-L test (Union14M-L-trained)accuracy: 82.41 -> 83.87Improved
add query-to-condition and query-to-image cross-attention stages (ablation)IGTR-AR · Union14M-L test (Union14M-L-trained)accuracy: 81.23 -> 84.86Improved
add question-to-condition and question-to-image stages (ablation)IGTR-AR · Union14M-L test (Union14M-L-trained)accuracy: 82.01 -> 84.86Improved
train 60 epochs instead of 20 with instruction sampling (ablation)IGTR-PR · Union14M-L test (Union14M-L-trained)accuracy: 82.51 -> 85.29Improved
train 60 epochs instead of 20 with PR instruction only (ablation)Vanilla-PR · Union14M-L test (Union14M-L-trained)accuracy: 76.14 -> 77.05Improved
scale image encoder from 24M to 40M (SVTR-L) (ablation)IGTR-PR · Union14M-L test (Union14M-L-trained)accuracy: 82.51 -> 83.60Improved
scale image encoder from 24M to 40M (SVTR-L) (ablation)Vanilla-PR · Union14M-L test (Union14M-L-trained)accuracy: 76.14 -> 76.23No clear effect
add targeted strengthening sampling for rare charactersIGTR-PR · CTR rare characters (1-10 occurrences)accuracy: 63.88 -> 67.47Improved
add targeted strengthening (TS)IGTR-AR · CTR Chinese benchmarkavg accuracy: 76.37 -> 77.17Improved

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