JTON's Zen Grid encoding reduces token counts for tabular data by 15-60% versus standard JSON while showing a net +0.3 pp accuracy gain and 100% syntactic validity across tested LLMs.
Oren Ben-Kiki, Clark Evans, and Ingy döt Net
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
JTON: A Token-Efficient JSON Superset with Zen Grid Tabular Encoding for Large Language Models
JTON's Zen Grid encoding reduces token counts for tabular data by 15-60% versus standard JSON while showing a net +0.3 pp accuracy gain and 100% syntactic validity across tested LLMs.