BURMESE-SAN creates the first comprehensive Burmese NLP benchmark with seven subtasks and shows architecture, representation, and instruction tuning outweigh model scale for performance.
Commercial models continue to achieve the strongest performance, but the gap with open- weight models is steadily narrowing as architec- tures and training strategies improve
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
-
BURMESE-SAN: Burmese NLP Benchmark for Evaluating Large Language Models
BURMESE-SAN creates the first comprehensive Burmese NLP benchmark with seven subtasks and shows architecture, representation, and instruction tuning outweigh model scale for performance.