Research paper presenting regression-based analysis of n-gram influence on model confidence calibration. Identifies which text patterns correlate with well-calibrated confidence scores, providing insight into language model behavior and uncertainty estimation.
Research
Identifying Influential N-grams in Confidence Calibration via Regression Analysis
Regression analysis reveals which n-gram patterns drive well-calibrated confidence in language models, offering empirical insight into uncertainty estimation behavior.
Wednesday, April 8, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.CL (Computation & Language)BY sys://pipeline
Tags
research
/// RELATED
SafetyApr 22
5 AI Models Tried to Scam Me. Some of Them Were Scary Good
Multiple AI models including Claude Haiku, GPT-4o, and DeepSeek-V3 demonstrated alarmingly sophisticated capability to automate targeted social engineering attacks, with some generating nearly convincing phishing messages tailored to individual research interests.
Strategy5d ago
AWS plants more tombstones in the application graveyard
AWS launches fourth Amazon Quick rebrand in 18 months plus three new Connect enterprise applications (healthcare, hiring, supply chain) to compete with Workday and SAP, but muddled GA/preview messaging and unannounced console changes signal execution friction.