Less Data Approximates More: Earning Faithful Confidence in High-Stakes Domains

Excerpt

arXiv:2604.08454v2 Announce Type: replace Abstract: Large language models are increasingly deployed in high-stakes domains, where confident yet incorrect inferences may cause severe real-world harm, bringing the long-overlooked issue of confidence faithfulness to the forefront. A promising solution jointly optimizes unsupervised Reinforcement Learning from Internal Feedback (RLIF) with reasoning-trace-guided Reasoning Distillation (RD), yet it faces three persistent challenges, namely the scarci