The paper introduces SAVE, a framework that uses on-policy feedback and the value function to self-supervise and improve reward models, significantly enhancing RLHF performance across multiple benchma…

This paper provides a mathematical framework for studying different policy learning problems and shows reductions between them.

The paper introduces the concept of policy-invisible violations in LLM agents and proposes Sentinel, a counterfactual graph simulation framework, which significantly improves policy enforcement accura…

The paper describes the 2026 MOASEI Competition, which evaluates multi-agent decision-making under open-system conditions in wildfire fighting, cybersecurity, and ride-sharing domains.

The paper introduces WIRE, a pipeline for diagnosing live intra-policy rule conflicts in LLM agents by identifying and testing specific rule pairs within a single prompt policy that can co-govern a re…

PolicyGapper is an LLM-based tool that automatically detects inconsistencies and omissions between a mobile app's Google Play Data Safety Section and its official Privacy Policy, identifying thousands…

The paper introduces POLICYGUARD, a sub-agent verifier that ensures policy adherence in LLM agents by providing actionable feedback for the next turn based on full conversation context and self-reason…

The paper introduces LedgerAgent, an inference-time method for tool-calling agents that maintains observed task states in a separate ledger and checks state-dependent policy constraints before tool ca…

This paper analyzes the effect of using stale rollouts in asynchronous GRPO and derives a stability condition.

The paper proposes S2L-PO, a framework that uses smaller, naturally diverse models as structured explorers to enhance the policy-level diversity and performance of larger language models during traini…

The paper introduces PROPARAG, an automated framework that autonomously assesses how well organizational cybersecurity policies comply with standard security controls, achieving high F1 scores on real…

The paper proposes a two-timescale governance framework using a multi-agent LLM to safely update and guide RL agents for SDN-IoT defense, significantly improving performance and stability under advers…

This paper provides non-asytotic sample complexity guarantees for the Navigate and Stop algorithm in online tabular Reinforcement Learning, identifying additional attributes that affect the overall sa…

The paper identifies and measures a critical failure mode where LLM agents violate policies by losing or corrupting directive-bearing state during the process of assembling the decision context, and p…

This paper introduces a stochastic differential equation approximation for linear Temporal Difference (TD) learning under Markovian noise, explaining the constant-stepsize error floor.

Proposed a novel experimental protocol for online controlled experiments to reduce variance and improve statistical power by exploiting policy overlap and using $Δ$-Off-Policy Estimation methods.

This paper analyzes 18 state-level AI committee reports to understand how policymakers discuss AI benefits and risks, comparing them to established taxonomy and HCI scholars' concerns.