LLM+AL, an LLM-to-BC+ pipeline with solver-feedback self-revision, outperforms direct LLMs and LLM-generated Python on complex action-reasoning puzzles, though it still needs a few manual corrections on most hard cases.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LLM+AL: Bridging Large Language Models and Action Languages for Complex Reasoning about Actions
LLM+AL, an LLM-to-BC+ pipeline with solver-feedback self-revision, outperforms direct LLMs and LLM-generated Python on complex action-reasoning puzzles, though it still needs a few manual corrections on most hard cases.