AI Paper Atlas

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

TDATR: Improving End-to-End Table Recognition via Table Detail-Aware Learning and Cell-Level Visual Alignment

Chunxia Qin, Chenyu Liu, Pengcheng Xia, Jun Du, Baocai Yin, Bing Yin, Cong Liu

CVPR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Layout-aware document understanding · 2 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 table content recognition tasks to detail-aware learning (ablation)TDATR · iFLYTAB-fullTEDS: 82.63 -> 91.57Improved
add table structure understanding tasks (ablation)TDATR · iFLYTAB-fullTEDS: 82.63 -> 90.44Improved
combine content and structure tasks (full two-stage) (ablation)TDATR · PubTabNetTEDS: 89.19 -> 94.38Improved
add content tasks on top of structure tasks (ablation)TDATR · iFLYTAB-fullTEDS: 90.44 -> 92.50Improved
add structure-guided refinement with multi-resolution features (ablation)SGCL · iFLYTAB-fullAP50: 89.22 -> 94.37Improved
add structure mask to bidirectional enhancement (ablation)SGCL · PubTabNetAP50: 90.02 -> 90.26Improved
add structure mask to bidirectional enhancement (ablation)SGCL · iFLYTAB-fullTEDS: 93.51 -> 93.41No clear effect
add cell localization module (C4) to HTML parsing (ablation)TDATR · PubTabNetTEDS: 94.38 -> 94.80Improved
use TDATR instead of Dolphin (same modeling approach)end-to-end TR · PubTables-1MTEDS: 95.48 -> 97.97Improved
use TDATR instead of TableFormer (24x more data)TSR · PubTabNet-ValTEDS-S: 96.75 vs 96.27Hurt
fine-tune two more epochs on PubTabNet (TDATR-ft) (ablation)TDATR · PubTabNet-ValTEDS: 95.12 -> 96.10Improved
use TDATR instead of PaddleOCR-VLtable recognition · OmniDocBench 1.5TEDS: 91.95 vs 87.96Hurt

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