作者位次 #4/4ACL 2026Findings· Findings of ACL
I ntro LM : Introspective Language Models via Prefilling-Time Self-Evaluation
Hossein Hosseini Kasnavieh · Gholamreza Haffari · Christopher Leckie · Adel N. Toosi
摘要
A major challenge for the operation of large language models (LLMs) is how to predict whether a specific LLM will produce sufficiently high-quality output for a given query. Existing approaches rely on external classifiers, most commonly BERT-based models, which suffer from limited context window...