AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

End-to-End Hyper-Relational Information Extraction for Engineering Diagrams via Dynamically Tokenized Relation Transformer

Tianyou Bai, Yan-Ming Zhang, Zixiang Zhang, Jibin Zhou, Fei Yin, Cheng-Lin Liu

CVPR 2026 · Graph learning › Knowledge and temporal graphs · approach: Graph-based dialog systems

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 denoising (DN) training (ablation)relation Transformer · P&IDAP50: 90.12 -> 93.94Improved
add relation-aware deformable decoupled visual attention (DDVA) (ablation)relation Transformer with DN · P&IDR@1000 (SGDET): 70.49 -> 77.65Improved
add contrastive denoising (CDN) (ablation)relation Transformer with DN and DDVA · P&IDR@1000 (SGDET): 77.65 -> 81.96Improved
add mixed query selection (MQS) (ablation)relation Transformer with DN, DDVA, CDN · P&IDR@1000 (SGDET): 81.96 -> 87.42Improved
add token scorers to prune 70% of tokens (ablation)Swin backbone · P&IDGFLOPs: 299.2 -> 90.5Improved
add token scorers only (no reconstruction branch) (ablation)Swin backbone · P&IDR@1000 (SGCLS): 87.99 -> 70.12Hurt
add reconstruction branch on top of scorers (ablation)token pruning · P&IDR@1000 (SGCLS): 70.12 -> 87.42Improved
add scorers only (ablation)Swin backbone · OCRCER: 1.21 -> 9.86Hurt
add reconstruction branch on top of scorers (ablation)token pruning · OCRCER: 9.86 -> 1.53Improved
replace two-stage Relationformer with one-stage DTRTscene graph generation · P&IDGFLOPs: 791.9 -> 90.5Improved
replace Relationformer with DTRTscene graph detection · P&IDR@1000: 76.43 -> 87.42Improved
replace RT-DETR with DTRT for entity detectionP&ID symbol detection · P&IDAP50: 99.27 -> 99.01No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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