LOGICPO: Efficient Translation of NL-based Logical Problems to FOL using LLMs and Preference Optimization

• URL: http://arxiv.org/abs/2506.18383v1
• Date: Mon, 23 Jun 2025 08:15:24 GMT
• Title: LOGICPO: Efficient Translation of NL-based Logical Problems to FOL using LLMs and Preference Optimization
• Authors: Koushik Viswanadha, Deepanway Ghosal, Somak Aditya,
• Abstract summary: We propose to use finetuning on a preference optimization dataset to learn to parse and represent a natural language problem as a whole to a consistent logical program.<n>Our best model with Phi-3.5 consistently outperforms GPT-3.5-turbo's by producing 10% more logically correct and with 14% less syntax errors.
• Score: 6.173941239083289
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Logical reasoning is a key task for artificial intelligence due to it's role in major downstream tasks such as Question Answering, Summarization. Recent methods in improving the reasoning ability of LLMs fall short in correctly converting a natural language reasoning problem to an equivalent logical formulation, which hinders the framework's overall ability to reason. Towards this, we propose to use finetuning on a preference optimization dataset to learn to parse and represent a natural language problem as a whole to a consistent logical program by 1) introducing a new supervised and preference optimization dataset LogicPO, and 2) adopting popular techniques such as Direct Preference Optimization (DPO), Kahneman-Tversky optimization (KTO) to finetune open-source LLMs. Our best model with Phi-3.5 consistently outperforms GPT-3.5-turbo's (8-shot) by producing 10% more logically correct and with 14% less syntax errors. Through the framework and our improved evaluation metrics, we offer a promising direction in improving the logical reasoning of LLMs by better representing them in their logical formulations.

Related papers

• Revisiting LLM Reasoning via Information Bottleneck [57.519119962528166]
  Large language models (LLMs) have recently demonstrated remarkable progress in reasoning capabilities through reinforcement learning with verifiable rewards (RLVR)<n>We present a theoretical characterization of LLM reasoning grounded in information bottleneck (IB) principle.<n>We propose IB-aware reasoning optimization (IBRO), a framework that encourages reasoning trajectories to be both informative about the final correct answer and generalizable.
  arXiv  Detail & Related papers  (2025-07-24T13:14:25Z)
• Don't Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models [68.96619605651155]
  Large reasoning models (LRMs) may drastically increase the output length due to overthinking.<n>We propose a dynamic optimization framework that segments model-generated reasoning paths into distinct thinking patterns.<n>Our method achieves up to a 12% accuracy improvement and reducing token usage from approximately 5,000 to 3,000 tokens.
  arXiv  Detail & Related papers  (2025-05-27T20:59:29Z)
• OptimAI: Optimization from Natural Language Using LLM-Powered AI Agents [8.441638148384389]
  We introduce OptimAI, a framework for solving Optimization problems described in natural language.<n>Our framework is built upon the following key roles: formulator, planner, coder and code critic.<n>Our approach attains 88.1% accuracy on the NLP4LP dataset and 82.3% on the Optibench dataset, reducing error rates by 58% and 52%, respectively, over prior best results.
  arXiv  Detail & Related papers  (2025-04-23T17:45:05Z)
• Teaching LLMs According to Their Aptitude: Adaptive Reasoning for Mathematical Problem Solving [55.895917967408586]
  Existing approaches to mathematical reasoning with large language models rely on Chain-of-Thought (CoT) for generalizability or Tool-Integrated Reasoning (TIR) for precise computation.<n>We propose TATA (Teaching LLMs According to Their Aptitude), an adaptive framework that enables LLMs to personalize their reasoning strategy spontaneously.
  arXiv  Detail & Related papers  (2025-02-17T16:56:23Z)
• Towards LLM-based optimization compilers. Can LLMs learn how to apply a single peephole optimization? Reasoning is all LLMs need! [0.9054540533394926]
  We show that OpenAI GPT-o1, despite not being fine-tuned, outperforms the fine-tuned Llama2 and GPT-4o.<n>Our findings indicate that this advantage is largely due to the chain-of-thought reasoning implemented in GPT-o1.
  arXiv  Detail & Related papers  (2024-12-11T18:44:31Z)
• HYBRIDMIND: Meta Selection of Natural Language and Symbolic Language for Enhanced LLM Reasoning [25.192089674713365]
  We introduce HYBRIDMIND, an adaptive strategy that selects the optimal reasoning approach for each reasoning problem.<n>We find that fine-tuning LLaMA-3.1-8B-Instruct as a meta-selector outperforms GPT-4o's natural language reasoning.
  arXiv  Detail & Related papers  (2024-09-28T15:12:55Z)
• Automated Theorem Provers Help Improve Large Language Model Reasoning [0.18416014644193066]
  We show how accuracy can be improved with a neuro-symbolic architecture. We define a framework of syntactic and semantic error categories. We extend our method with capabilities for automatically correcting syntactic and semantic errors.
  arXiv  Detail & Related papers  (2024-08-07T01:03:56Z)
• LOGIC-LM++: Multi-Step Refinement for Symbolic Formulations [1.024113475677323]
  This paper proposes Logic-LM++, an improvement on Logic-LM. It uses the ability of LLMs to do pairwise comparisons, allowing the evaluation of the refinements suggested by the LLM.
  arXiv  Detail & Related papers  (2024-06-22T12:50:41Z)
• LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models [63.14196038655506]
  We introduce LogicAsker, a novel approach for evaluating and enhancing the logical reasoning capabilities of large language models (LLMs) Our methodology reveals significant gaps in LLMs' learning of logical rules, with identified reasoning failures ranging from 29% to 90% across different models. We leverage these findings to construct targeted demonstration examples and fine-tune data, notably enhancing logical reasoning in models like GPT-4o by up to 5%.
  arXiv  Detail & Related papers  (2024-01-01T13:53:53Z)
• LINC: A Neurosymbolic Approach for Logical Reasoning by Combining Language Models with First-Order Logic Provers [60.009969929857704]
  Logical reasoning is an important task for artificial intelligence with potential impacts on science, mathematics, and society. In this work, we reformulating such tasks as modular neurosymbolic programming, which we call LINC. We observe significant performance gains on FOLIO and a balanced subset of ProofWriter for three different models in nearly all experimental conditions we evaluate.
  arXiv  Detail & Related papers  (2023-10-23T17:58:40Z)
• SatLM: Satisfiability-Aided Language Models Using Declarative Prompting [68.40726892904286]
  We propose a new satisfiability-aided language modeling (SatLM) approach for improving the reasoning capabilities of large language models (LLMs) We use an LLM to generate a declarative task specification rather than an imperative program and leverage an off-the-shelf automated theorem prover to derive the final answer. We evaluate SATLM on 8 different datasets and show that it consistently outperforms program-aided LMs in the imperative paradigm.
  arXiv  Detail & Related papers  (2023-05-16T17:55:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.