C. Koechli

Evidence Use and Identifier-Conditioned Prior Knowledge in Large Language Model Classification of Oncology Trials Assessed Through Progressive Content Removal and Counterfactual Testing: Comparative Analysis

Testing whether oncology randomized trial success classification is driven by abstract evidence or by identifier-conditioned prior knowledge, and whether models follow counterfactual outcome evidence when it conflicts with original trial identifiers showed that identifiers can carry predictive signal and occasionally c...