Codesota · Models · TFLOPUpstage AI2 results · 1 benchmarks
Model card

TFLOP.

Upstage AIopen-sourceUnknown paramsLayout Pointer mechanism; span-aware contrastive supervision; reformulates TSR as text region pointing

IJCAI 2024. Achieves SOTA on PubTabNet, FinTabNet, SynthTabNet. Smallest TEDS/TEDS-Struct gap among compared methods.

§ 01 · Benchmarks

Every benchmark TFLOP has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01pubtabnetComputer Vision · Table Recognitionteds-struct98.4%#1/142025-01-21source ↗
02pubtabnetComputer Vision · Table Recognitionteds-all-samples96.7%#6/162025-01-21source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where TFLOP actually performs.

Computer Vision
1
benchmark
avg rank #3.5
§ 03 · Papers

1 paper with results for TFLOP.

  1. 2025-01-21· Computer Vision· 2 results

    TFLOP: Table Structure Recognition Framework with Layout Pointer Mechanism

§ 05 · Sources & freshness

Where these numbers come from.

arxiv
2
results
2 of 2 rows marked verified.