Addressing Logical Fallacies In Scientific Reasoning From Large Language Models: Towards a Dual-Inference Training Framework

TL;DR

This paper identifies weaknesses in large language models' scientific reasoning, especially with negation and faulty premises, and proposes a dual-inference training framework that enhances their robustness and interpretability by integrating affirmative and negation-aware reasoning.

Contribution

It introduces a novel dual-reasoning training paradigm grounded in formal logic and cognitive science, improving LLMs' ability to reject invalid inferences and resist logical fallacies.

Findings

Existing LLMs show systematic reasoning weaknesses in scientific domains.

The proposed dual-inference framework improves model robustness and interpretability.

Models trained with this method better reject invalid inferences and handle negation.

Abstract

Large Language Models (LLMs) have transformed natural language processing and hold growing promise for advancing science, healthcare, and decision-making. Yet their training paradigms remain dominated by affirmation-based inference, akin to \textit{modus ponens}, where accepted premises yield predicted consequents. While effective for generative fluency, this one-directional approach leaves models vulnerable to logical fallacies, adversarial manipulation, and failures in causal reasoning. This paper makes two contributions. First, it demonstrates how existing LLMs from major platforms exhibit systematic weaknesses when reasoning in scientific domains with negation, counterexamples, or faulty premises \footnote{Code to recreate these experiments are at https://github.com/hannahdavidsoncollege-maker/ScientificReasoningForEnvironment-MedicineWithLLMs. Second, it introduces a dual-reasoning training framework that integrates affirmative generation with structured counterfactual denial. Grounded in formal logic, cognitive science, and adversarial training, this training paradigm formalizes a computational analogue of ``denying the antecedent'' as a mechanism for disconfirmation and robustness. By coupling generative synthesis with explicit negation-aware objectives, the framework enables models that not only affirm valid inferences but also reject invalid ones, yielding systems that are more resilient, interpretable, and aligned with human reasoning.