SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.18357 · ACL · 2024

Faithful Logical Reasoning via Symbolic Chain-of-Thought

Qian Liu, Hao Fei, Liangming Pan, Jundong Xu, Mong Li Lee, W. Hsu

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 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
Aiden0526/SymbCoT canonical 7 of 7
FunctionStatusWhere it lives
argmax Ran Aiden0526/SymbCoT/baselines/evaluation.py
code served (permissive licence) · get_code("c336622e652dd934")
compute_f1_score Ran Aiden0526/SymbCoT/evaluate.py
code served (permissive licence) · get_code("8f1a04852f7b3744")
dispatch_openai_chat_requests Ran Aiden0526/SymbCoT/utils.py
code served (permissive licence) · get_code("4e1c4ff342f96bc9")
dispatch_openai_prompt_requests Ran Aiden0526/SymbCoT/utils.py
code served (permissive licence) · get_code("94928eafb9afc8c7")
evaluate_performance Ran Aiden0526/SymbCoT/evaluate.py
code served (permissive licence) · get_code("175a1319fe2df65c")
extract_number Ran Aiden0526/SymbCoT/baselines/evaluation.py
code served (permissive licence) · get_code("471cfa8084196a13")
normalize_text Ran Aiden0526/SymbCoT/baselines/evaluation.py
code served (permissive licence) · get_code("71986ed6ac7b9157")

Repositories linked to this paper

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

Abstract

While the recent Chain-of-Thought (CoT) technique enhances the reasoning ability of large language models (LLMs) with the theory of mind, it might still struggle in handling logical reasoning that relies much on symbolic expressions and rigid deducing rules. To strengthen the logical reasoning capability of LLMs, we propose a novel Symbolic Chain-of-Thought, namely SymbCoT, a fully LLM-based framework that integrates symbolic expressions and logic rules with CoT prompting. Technically, building upon an LLM, SymbCoT 1) first translates the natural language context into the symbolic format, and then 2) derives a step-by-step plan to solve the problem with symbolic logical rules, 3) followed by a verifier to check the translation and reasoning chain. Via thorough evaluations on 5 standard datasets with both First-Order Logic and Constraint Optimization symbolic expressions, SymbCoT shows striking improvements over the CoT method consistently, meanwhile refreshing the current stateof-the-art performances. We further demonstrate that our system advances in more faithful, flexible, and explainable logical reasoning. To our knowledge, this is the first to combine symbolic expressions and rules into CoT for logical reasoning with LLMs. Code is open at https://github.com/Aiden0526/SymbCoT.

For agents

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

get_harvested_code_for_paper("2405.18357")
get_code_for_paper("2405.18357")
have("2405.18357")

Connect an agent — have() is free.