Nous: Learning and Certifying Memory Decisions Before Source Calibration

arXiv:2610.00094 · cs.LG, cs.AI · Submitted 2026-09-08 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-08

Updated: 2026-10-05

Terminology

• Calibration — 1× in this paper · explained in 7 episodes

Sources

• When Does Belief-Based Agent Memory Help? Reliability-Conditional Updating and Provenance-Capped Poisoning Defense
• Fundamental limits for learning hidden Markov model parameters
• MemTX: Transactional Belief Commit for Stateful Agent Memory
• MemTxn: A Transaction Boundary for Source-Supported Updates and Complete-State Recovery in Agent Memory
• Classification with unknown class-conditional label noise on non-compact feature spaces
• Noise tolerance of learning to rank under class-conditional label noise
• Learning from Label Proportion with Online Pseudo-Label Decision by Regret Minimization

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks