MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables
This work addresses the need for better benchmarks to assess complex reasoning in LLMs, though it is incremental as it builds on existing literature-based evaluation frameworks.
The authors tackled the problem of evaluating abstract moral reasoning in large language models (LLMs) by introducing MORABLES, a benchmark based on fables and short stories, and found that while larger models perform better, they are brittle and rely on superficial patterns, with the best models contradicting themselves in roughly 20% of cases.
As LLMs excel on standard reading comprehension benchmarks, attention is shifting toward evaluating their capacity for complex abstract reasoning and inference. Literature-based benchmarks, with their rich narrative and moral depth, provide a compelling framework for evaluating such deeper comprehension skills. Here, we present MORABLES, a human-verified benchmark built from fables and short stories drawn from historical literature. The main task is structured as multiple-choice questions targeting moral inference, with carefully crafted distractors that challenge models to go beyond shallow, extractive question answering. To further stress-test model robustness, we introduce adversarial variants designed to surface LLM vulnerabilities and shortcuts due to issues such as data contamination. Our findings show that, while larger models outperform smaller ones, they remain susceptible to adversarial manipulation and often rely on superficial patterns rather than true moral reasoning. This brittleness results in significant self-contradiction, with the best models refuting their own answers in roughly 20% of cases depending on the framing of the moral choice. Interestingly, reasoning-enhanced models fail to bridge this gap, suggesting that scale - not reasoning ability - is the primary driver of performance.