作者位次 #4/5ACL 2026Poster· Natural Language Processing
R e T race QA : Evaluating Reasoning Traces of Small Language Models in Commonsense Question Answering
Francesco Maria Molfese · Luca Moroni · Ciro Porcaro · Simone Conia · Roberto Navigli
摘要
While Small Language Models (SLMs) have demonstrated promising performance on an increasingly wide array of commonsense reasoning benchmarks, current evaluation practices rely almost exclusively on the accuracy of their final answers, neglecting the validity of the reasoning processes that lead t...