Page 110 of 151

AllHigh signalRecent
6029 articles
arXiv cs.AI·

Operational AI Deployment Assurance: Governance-State Orchestration Under Threshold-Sensitive Deployment Conditions -- A Governance Framework for High-Stakes AI Systems

OADA is a governance framework for high-stakes AI systems that translates fairness metric instability, threshold sensitivity, and operational uncertainty into deployment-oriented assurance decisions. Tested on facial recognition and healthcare, it introduces Deployment Assurance Scores, escalation states, and Threshold Stability Zones to actively govern deployment readiness rather than rely on post-hoc auditing.

AI safetyAlignmentEvals
SIG
62
HYP
28
arXiv cs.LG·

Energy-Gated Attention and Wavelet Positional Encoding: Complementary Inductive Biases for Transformer Attention

Two complementary mechanisms improve transformer attention: Energy-Gated Attention (EGA) selects informative tokens via linear projection; Morlet Positional Encoding (MoPE) replaces sinusoidal encodings with learned Gaussian wavelets. On TinyShakespeare, their combination achieves +0.119 validation loss improvement, exceeding the sum of individual parts.

PapersReasoning
SIG
62
HYP
25