0
OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones?
Zijian Chen, Tingzhu Chen, Wenjun Zhang et al., 2025.
What it's used for here
Benchmark of 5,523 images and five tasks (recognition, rejoining, classification, retrieval, deciphering) on 23 large multimodal models; the best models match untrained humans on deciphering and trail them on fine perception. ICLR 2025.
Our copy
Research copy, for editors and curators. The work is the publisher's, so readers are sent to it. We keep this copy so the citation can be checked if the page changes or goes.
arxiv.org · fetched 31 Oct 2026
Loading…
Cited by on this site 3 records
Cite it
Zijian Chen, Tingzhu Chen, Wenjun Zhang, & Guangtao Zhai (2025). OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones?. https://doi.org/10.48550/arXiv.2412.01175
@inproceedings{chen2025obibench,
author = {Zijian Chen} and {Tingzhu Chen} and {Wenjun Zhang} and {Guangtao Zhai},
title = {OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones?},
year = {2025},
doi = {10.48550/arXiv.2412.01175},
url = {https://arxiv.org/abs/2412.01175}
}TY - GEN TI - OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones? AU - Zijian Chen AU - Tingzhu Chen AU - Wenjun Zhang AU - Guangtao Zhai PY - 2025 DO - 10.48550/arXiv.2412.01175 UR - https://arxiv.org/abs/2412.01175 ER -
et al.. “OBI-Bench: Can LMMs Aid in Study of Ancient Script on Oracle Bones?.” 2025. https://doi.org/10.48550/arXiv.2412.01175.