Policy-Verified Self-Training: Learning Agent Compliance from Verifiable Policy Rewards
cs.LG, cs.AI
Submitted: 2026-10-14
Updated: 2026-10-14
Comments: 16 pages, 3 figures, 13 tables. Companion paper: arXiv:2610.11030
Code: https://github.com/iamupd/NOMOS
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- StruQ: Defending Against Prompt Injection with Structured Queries
- SecAlign: Defending Against Prompt Injection with Preference Optimization
- Meta-SecAlign: Training LLMs against Prompt Injection for Robust Agents
- AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
- Defeating Prompt Injections by Design
- AgentLTL: A Trace-Verification Framework for Measuring, Enforcing, and Training Procedural Compliance in Tool-Using LLM Agents
- Reinforced Self-Training (ReST) for Language Modeling
- LoRA: Low-Rank Adaptation of Large Language Models
- RLVP: Penalize the Path, Reward the Outcome
- Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Progent: Securing AI Agents with Privilege Control
- The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions
- AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- NOMOS: Compiling Written Policies into Statically Verified Tool-Call Gates for LLM Agents
- Scaling Relationship on Learning Mathematical Reasoning with Large Language Models
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