From fc7c564988a1f8d083c73b3344ed2803920bc0cc Mon Sep 17 00:00:00 2001 From: Kung-hsiang Date: Tue, 4 Jun 2024 20:53:15 -0700 Subject: [PATCH] update acl 2024 --- README.md | 19 +++++++++++-------- docs/index.html | 8 ++++---- 2 files changed, 15 insertions(+), 12 deletions(-) diff --git a/README.md b/README.md index 4b7b21c..a258354 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,4 @@ -# Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning +# [ACL 2024 Findings] Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning
Kung-Hsiang Huang†, Mingyang Zhou*, Hou Pong Chan‡, @@ -91,7 +91,7 @@ Results are shown in the below figure and table. We found that all captioning mo - [ ] Evaluation scripts -## The CHOCOLATE Benchmark +## The CHOCOLATE Benchmark We release the data for the CHOCOLATE benchmark at `data/chocolate.json`. CHOCOLATE is also available on [HuggingFace](https://huggingface.co/datasets/khhuang/CHOCOLATE)🤗. @@ -155,7 +155,7 @@ The meta-evaluation scripts can be found in [ChartVE Meta-evaluation.ipynb](http The proposed C2TFEC framework consists of two components: chart-to-table conversion and table-based error rectification. -### Chart-To-Table +### Chart-To-Table The Chart-To-Table model ([khhuang/chart-to-table](https://huggingface.co/khhuang/chart-to-table)) is trained to convert a chart into a structured table. The generated tables use &&& to delimit rows and | to delimit columns. The underlying architecture of this model is UniChart. Below, we provide an example of how to use our Chart-To-Table model. @@ -208,7 +208,7 @@ extracted_table = sequence.split("")[1].strip() ## Citation ```bibtex -@misc{huang-etal-2023-do, +@inproceedings{huang-etal-2024-lvlms, title = "Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning", author = "Huang, Kung-Hsiang and Zhou, Mingyang and @@ -218,10 +218,13 @@ extracted_table = sequence.split("")[1].strip() Zhang, Lingyu and Chang, Shih-Fu and Ji, Heng", - year={2023}, - eprint={2312.10160}, - archivePrefix={arXiv}, - primaryClass={cs.CL} + booktitle = "Findings of the Association for Computational Linguistics: ACL 2024", + month = aug, + year = "2024", + publisher = "Association for Computational Linguistics", + url = "https://aclanthology.org/2023.findings-acl.85", + doi = "10.18653/v1/2023.findings-acl.85", + pages = "1314--1326", } ``` diff --git a/docs/index.html b/docs/index.html index 66012e4..767ae4a 100644 --- a/docs/index.html +++ b/docs/index.html @@ -367,7 +367,7 @@

Leaderboard on Factual Inconsistency Det Link -0.014 0.105 - 0.291 + 0.291 @@ -386,13 +386,13 @@

Leaderboard on Factual Inconsistency Det Large Vision-language Model Link 0.157 - 0.205 + 0.205 0.215 - GPT-4O + GPT-4o Large Vision-language Model Link 0.250 @@ -416,7 +416,7 @@

Leaderboard on Factual Inconsistency Det ChartVE Small Vision-language Model Link - 0.178 + 0.178 0.091 0.215