Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

TableSeq: Unified Generation of Structure, Content, and Layout

About

We present TableSeq, an image-only, end-to-end framework for joint table structure recognition, content recognition, and cell localization. The model formulates these tasks as a single sequence-generation problem: one decoder produces an interleaved stream of \texttt{HTML} tags, cell text, and discretized coordinate tokens, thereby aligning logical structure, textual content, and cell geometry within a unified autoregressive sequence. This design avoids external OCR, auxiliary decoders, and complex multi-stage post-processing. TableSeq combines a lightweight high-resolution FCN-H16 encoder with a minimal structure-prior head and a single-layer transformer encoder, yielding a compact architecture that remains effective on challenging layouts. Across standard benchmarks, TableSeq achieves competitive or state-of-the-art results while preserving architectural simplicity. It reaches 95.23 TEDS / 96.83 S-TEDS on PubTabNet, 97.45 TEDS / 98.69 S-TEDS on FinTabNet, and 99.79 / 99.54 / 99.66 precision / recall / F1 on SciTSR under the CAR protocol, while remaining competitive on PubTables-1M under GriTS. Beyond TSR/TCR, the same sequence interface generalizes to index-based table querying without task-specific heads, achieving the best IRDR score and competitive ICDR/ICR performance. We also study multi-token prediction for faster blockwise decoding and show that it reduces inference latency with only limited accuracy degradation. Overall, TableSeq provides a practical and reproducible single-stream baseline for unified table recognition, and the source code will be made publicly available at https://github.com/hamdilaziz/TableSeq.

Laziz Hamdi, Amine Tamasna, Pascal Boisson, Thierry Paquet• 2026

Related benchmarks

TaskDatasetResultRank
Table RecognitionPubTabNet (test)
TEDS (Simple)96.83
70
Table Structure RecognitionFinTabNet
S-TEDS98.69
40
Table Structure RecognitionSciTSR
Precision99.79
18
Table Structure RecognitionPubTables-1M
GriTS Top Score99.1
15
Row Index-based Data Recognition (IRDR)SciTSR + PubTabNet 1,500-image (test and val)
F1 Score67.73
7
Index-based Cell Recognition (ICR)SciTSR + PubTabNet 1,500-image (test val)
Accuracy42.85
7
Column Index-based Data Recognition (ICDR)SciTSR + PubTabNet 1,500-image (test val)
F1 Score67.5
7
Cell content bounding-box detectionPubTabNet (val)
AP@5096.5
5
Table Structure RecognitionICDAR 2013 (test)
Precision95.78
5
Showing 9 of 9 rows

Other info

Follow for update