Enhancing Syllogistic Reasoning in Biomedical NLI: Key Insights and Challenges

Table of Links

1. Abstract and Introduction
2. SylloBio-NLI
3. Empirical Evaluation
4. Related Work
5. Conclusions
6. Limitations and References

A. Formalization of the SylloBio-NLI Resource Generation Process

B. Formalization of Tasks 1 and 2

C. Dictionary of gene and pathway membership

D. Domain-specific pipeline for creating NL instances and E Accessing LLMs

F. Experimental Details

G. Evaluation Metrics

H. Prompting LLMs - Zero-shot prompts

I. Prompting LLMs - Few-shot prompts

J. Results: Misaligned Instruction-Response

K. Results: Ambiguous Impact of Distractors on Reasoning

L. Results: Models Prioritize Contextual Knowledge Over Background Knowledge

M Supplementary Figures and N Supplementary Tables

5 Conclusions

In this work, we proposed a novel methodological framework, SylloBio-NLI, designed to evaluate the syllogistic reasoning capabilities of state-of-the-art LLMs within the biomedical domain. Through comprehensive analysis across 28 syllogistic schemes, we assessed the performance of eight different models under varying conditions, including zero-shot and few-shot settings. Our results show that both techniques exhibit high sensitivity to superficial lexical variations, highlighting a dependency between reliability, models’ architecture, and pre-training regime. Overall, our evaluation indicates that, while few-shot strategies have the potential to elicit syllogistic reasoning in LLMs, existing models are still far from achieving the robustness and consistency required for safe biomedical NLI applications.