SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.00830 · AAAI · 2025

LLM+AL: Bridging Large Language Models and Action Languages for Complex Reasoning About Actions

Joohyung Lee, Adam Ishay

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 12 functions out of this paper's own repositories and ran 7 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
azreasoners/llm-al — 7 of 12
FunctionStatusWhere it lives
clean_query Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("511e7db04ad54b55")
clean_signature Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("1544955107ceb787")
get_response Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("535397c1d5ef3d19")
get_response_check Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("06a9520971a24332")
split_prog Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("506b4f660abe25d4")
split_prog_query Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("f3ebc1f0736854c2")
split_sample_queries Ran azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("71bf09ae60d68be5")
LLM Not yet run azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("15771455409e6f78")
extract_signature2 Not yet run azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("0503e83a6745eb90")
save_cache Not yet run azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("82c955ece882d449")
save_cache_basic Not yet run azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("8505e8e8b6a6a2f6")
write_intermediate3 Not yet run azreasoners/llm-al/utils.py
pointer only (licence: NONE) · get_code("6b79906f2d136713")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Large Language Models (LLMs) have made significant strides in various intelligent tasks but still struggle with complex action reasoning tasks that require systematic search. To address this limitation, we propose a method that bridges the natural language understanding capabilities of LLMs with the symbolic reasoning strengths of action languages. Our approach, termed LLM+AL, leverages the LLM's strengths in semantic parsing and commonsense knowledge generation alongside the action language's proficiency in automated reasoning based on encoded knowledge. We compare LLM+AL against state-of-the-art LLMs, including CHATGPT-4, CLAUDE 3 OPUS, GEMINI ULTRA 1.0, and O1-PREVIEW, using benchmarks for complex reasoning about actions. Our findings indicate that, although all methods exhibit errors, LLM+AL, with relatively minimal human corrections, consistently leads to correct answers, whereas standalone LLMs fail to improve even with human feedback. LLM+AL also contributes to automated generation of action languages.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2501.00830")
get_code_for_paper("2501.00830")
have("2501.00830")

Connect an agent — have() is free.