Chart-CoCa makes a vision language model improve its own chart question answering by generating synthetic charts via code, extracting exact answers from that code, and training itself to synthesize a final answer from its own candidate responses.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Chart-CoCa: Self-Improving Chart Understanding of Vision LMs via Code-Driven Synthesis and Candidate-Conditioned Answering
Chart-CoCa makes a vision language model improve its own chart question answering by generating synthetic charts via code, extracting exact answers from that code, and training itself to synthesize a final answer from its own candidate responses.