A new benchmark and glyph-driven fine-tuning framework (GEVO) improve multimodal LLMs' performance on analyzing ancient Chinese character evolution across 11 tasks, with gains shown even for 2B-scale models.
Bronze Inscription → Regular Script → Oracle Bone Script,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Enhancing Multimodal Large Language Models for Ancient Chinese Character Evolution Analysis via Glyph-Driven Fine-Tuning
A new benchmark and glyph-driven fine-tuning framework (GEVO) improve multimodal LLMs' performance on analyzing ancient Chinese character evolution across 11 tasks, with gains shown even for 2B-scale models.