作者位次 #3/5ICLR 2026Poster· Social Aspects / Fairness, Equity, Justice and Safety
Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
James Oldfield · Philip Torr · Ioannis Patras · Adel Bibi · Fazl Barez
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
James Oldfield · Philip Torr · Ioannis Patras · Adel Bibi · Fazl Barez
Zhaohan Zhang · Ziquan Liu · Ioannis Patras
Assessing the reliability of Large Language Models (LLMs) by confidence elicitation is a prominent approach to AI safety in high-stakes applications, such as healthcare and finance. Existing methods either require expensive computational overhead or suffer from poor calibration, making them impra...
Zhaohan Zhang · Chengzhengxu Li · Xiaoming Liu · Chao Shen · Ziquan Liu · Ioannis Patras
Large Language Models (LLMs) are increasingly applied in high-stakes domains such as finance, healthcare, and education, where reliable multi-turn interactions with users are essential. However, existing work on confidence estimation and calibration, a major approach to building trustworthy LLM s...