Reproduce “Logics-Parsing-Omni Technical Report” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xin An, Jingyi Cai, Xiangyang Chen, Huayao Liu, Peiting Liu, Peng Wang, Bei Yang, Xiuwen Zhu and 17 more
Logics-Parsing-Omni integrates multimodal parsing by progressively locating objects, text, and events, recognizing contents and attributes, and interpreting them together. It produces structured evidence intended to remain locatable, enumerable, and traceable. In graphics, Parsing & Caption raises chart Logical Reasoning from 73.97 to 90.87. Broader benchmarks show gains on Graphics and Text-Rich Video, but not Natural Image or Natural Video.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.