A study on the Gemma 3 4B model investigated methods to improve its verbal confidence in responses. Initial attempts using a filtered dataset for confidence-conditioned supervised fine-tuning (CSFT) yielded negative results, decreasing performance. However, an exploratory approach that removed the filter and trained on all calibration items significantly improved the model's ability to predict verbal correctness, achieving an AUROC2 of 0.774 on TriviaQA. AI
影响 Demonstrates a potential method to improve confidence calibration in smaller LLMs, impacting their reliability in downstream applications.
排序理由 This is a research paper detailing experimental results on a specific model's performance.
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
