The authors create a novel dataset and show that fine-tuned small open-weight LLMs can translate natural language to ATL/ATL* specifications with semantic accuracy (0.84) statistically matching strong few-shot proprietary baselines (0.86).
arXiv preprint arXiv:2502.16339 (2025)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.MA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Translating Natural Language to Strategic Temporal Specifications via LLMs
The authors create a novel dataset and show that fine-tuned small open-weight LLMs can translate natural language to ATL/ATL* specifications with semantic accuracy (0.84) statistically matching strong few-shot proprietary baselines (0.86).