The Refutability Gap: Challenges in Validating Reasoning by Large Language Models

TL;DR

This paper critiques the scientific validity of claims about LLMs' reasoning abilities, emphasizing the need for falsifiability, transparency, and reproducibility to ensure credible progress in AI research.

Contribution

It identifies methodological pitfalls in current LLM reasoning research and proposes guidelines for transparency and reproducibility to improve scientific rigor.

Findings

Current LLM reasoning claims lack falsifiability.

Opaque training data hampers verification of novelty.

Absence of counterfactuals biases evaluation.

Abstract

Recent reports claim that Large Language Models (LLMs) have achieved the ability to derive new science and exhibit human-level general intelligence. We argue that such claims are not rigorous scientific claims, as they do not satisfy Popper's refutability principle (often termed falsifiability), which requires that scientific statements be capable of being disproven. We identify several methodological pitfalls in current AI research on reasoning, including the inability to verify the novelty of findings due to opaque and non-searchable training data, the lack of reproducibility caused by continuous model updates, and the omission of human-interaction transcripts, which obscures the true source of scientific discovery. Additionally, the absence of counterfactuals and data on failed attempts creates a selection bias that may exaggerate LLM capabilities. To address these challenges, we propose guidelines for scientific transparency and reproducibility for research on reasoning by LLMs. Establishing such guidelines is crucial for both scientific integrity and the ongoing societal debates regarding fair data usage.