Learning Label Dependencies for Visual Information Extraction

Authors: Minghong Yao, Liansheng Zhuang, Houqiang Li, Jiuchang Wei

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on public datasets have demonstrated the effectiveness of our framework.
Researcher Affiliation Academia University of Science and Technology of China
Pseudocode Yes Algorithm 1 Training procedures.
Open Source Code No The paper does not contain an explicit statement about the release of open-source code or a link to a code repository for the described methodology.
Open Datasets Yes Table 1 lists three VIE benchmark datasets: FUNSD [Jaume et al., 2019], CORD [Park et al., 2019], and SROIE [Huang et al., 2019].
Dataset Splits No Table 1 lists three VIE benchmark datasets: FUNSD Train 149, Test 50 Forms CORD Train 800, Test 100 Receipts SROIE Train 626, Test 276 Receipts.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts) used for running the experiments.
Software Dependencies No The paper mentions using the 'Adam W optimizer' but does not specify version numbers for any software dependencies or libraries.
Experiment Setup Yes The Adam W optimizer is employed for fine-tuning, with an initial learning rate of 5e-5 and a linear decay learning rate scheduler. We employed the Adam W optimizer to train it, with a learning rate set to 5e-5 and a batch size of 1. For the fixed-point search, we utilized the Broyden [Broyden, 1965] method, with a maximum iteration step limit set to 10. We set Tinner to be 4 during training and testing.