SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.02623 · 2022

Unifying Vision, Text, and Layout for Universal Document Processing

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 17 functions out of this paper's own repositories and ran 14 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
DS4SD/MarkushGrapher pwc_unofficial 11 of 14
copy not recorded — 3 of 3
FunctionStatusWhere it lives
custom_huber Ran DS4SD/MarkushGrapher/markushgrapher/core/trainers/losses.py
code served (permissive licence) · get_code("f587be8f27266175")
custom_huber2 Ran DS4SD/MarkushGrapher/markushgrapher/core/trainers/losses.py
code served (permissive licence) · get_code("8ad08bf63e399fe5")
default Ran this paper's copy was not recorded; identical code first harvested from ashen-sensored/sd_webui_SAG
pointer only · get_code("424012cb37b31172")
estimate_word_width Ran DS4SD/MarkushGrapher/markushgrapher/core/common/data_preprocessing.py
code served (permissive licence) · get_code("68402efebc760365")
exists Ran this paper's copy was not recorded; identical code first harvested from ThomasMrY/VCT
pointer only · get_code("aa5486a3650902d8")
get_last_checkpoint Ran DS4SD/MarkushGrapher/markushgrapher/core/common/utils.py
code served (permissive licence) · get_code("9f1130d32c6430f8")
get_linear_with_fact_schedule_with_warmup Ran DS4SD/MarkushGrapher/markushgrapher/core/trainers/optimization.py
code served (permissive licence) · get_code("a3b662bbc3a23327")
get_scheduler Ran DS4SD/MarkushGrapher/markushgrapher/core/trainers/optimization.py
code served (permissive licence) · get_code("e78d0e6d79459413")
huber_loss Ran DS4SD/MarkushGrapher/markushgrapher/core/trainers/losses.py
code served (permissive licence) · get_code("a83a110a9551382d")
pad_sequence_native Ran DS4SD/MarkushGrapher/markushgrapher/core/trainers/data_collator.py
code served (permissive licence) · get_code("4d78a796062fd89b")
setup_logging Ran DS4SD/MarkushGrapher/markushgrapher/core/common/begin.py
code served (permissive licence) · get_code("fbd45e87bae891a7")
split_bounding_box_for_words Ran DS4SD/MarkushGrapher/markushgrapher/core/common/data_preprocessing.py
code served (permissive licence) · get_code("084ee3628fc5d124")
split_sentence_into_words Ran DS4SD/MarkushGrapher/markushgrapher/core/common/data_preprocessing.py
code served (permissive licence) · get_code("32b4122ad5350b54")
uniq Ran this paper's copy was not recorded; identical code first harvested from JingWu321/MUNBa
pointer only · get_code("9a299fe5ae09e407")
calculate_iou Not yet run DS4SD/MarkushGrapher/markushgrapher/core/common/utils.py
code served (permissive licence) · get_code("ea1bb4ac7f48a7b5")
clamp Not yet run DS4SD/MarkushGrapher/markushgrapher/core/common/utils.py
code served (permissive licence) · get_code("59eacee354fa7772")
last_checkpoint Not yet run DS4SD/MarkushGrapher/markushgrapher/core/common/begin.py
code served (permissive licence) · get_code("5b942743f6a5251f")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We propose Universal Document Processing (UDOP), a foundation Document AI model which unifies text, image, and layout modalities together with varied task formats, including document understanding and generation. UDOP leverages the spatial correlation between textual content and document image to model image, text, and layout modalities with one uniform representation. With a novel Vision-Text-Layout Transformer, UDOP unifies pretraining and multi-domain downstream tasks into a prompt-based sequence generation scheme. UDOP is pretrained on both large-scale unlabeled document corpora using innovative self-supervised objectives and diverse labeled data. UDOP also learns to generate document images from text and layout modalities via masked image reconstruction. To the best of our knowledge, this is the first time in the field of document AI that one model simultaneously achieves high-quality neural document editing and content customization. Our method sets the state-of-the-art on 8 Document AI tasks, e.g., document understanding and QA, across diverse data domains like finance reports, academic papers, and websites. UDOP ranks first on the leaderboard of the Document Understanding Benchmark.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2212.02623")
get_code_for_paper("2212.02623")
have("2212.02623")

Connect an agent — have() is free.