arXiv Machine Learning June 2, 2026 By Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma

Auditing Near-Optimal Policies Can Be Exponentially Hard: Conditional Query Lower Bounds via Occupancy Rashomon Capacity

Read the original on arXiv Machine Learning →

arXiv:2606. 00414v1 Announce Type: new Abstract: When many reinforcement-learning policies achieve near-optimal return, a post-hoc auditor may have to distinguish among many behaviorally distinct but return-equivalent policies.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jul 7

How Much of the Routing Gap Is Real? Decomposing the Router-to-Oracle Gap into Reproducible Specialist Advantage and Single-Draw Label Noise

arXiv:2607. 03436v1 Announce Type: new Abstract: Routing among large language models (LLMs) promises better quality at lower cost, motivated by the reported gap between learned routers and a per-instance oracle.

By Teng-Ruei Chen

llms benchmarks

arXiv Machine Learning

Jun 25

Minimax PAC Bounds for Learning in Exogenous Contextual MDPs

arXiv:2606. 25170v1 Announce Type: cross Abstract: We study PAC learning in tabular discounted Markov decision processes with exogenous i.

By Corentin Pla, Hugo Richard, Marc Abeille, Vianney Perchet

agents reinforcement-learning

arXiv Machine Learning

Aug 10

Sub-Quadratic Bisimulation Metrics via Approximate Nearest Neighbors: Coverage-Augmented Guarantees and Computable Two-Sided Certificates

arXiv:2608. 06762v1 Announce Type: new Abstract: Bisimulation metrics quantify behavioral similarity in Markov decision processes, but their Wasserstein fixed-point operator updates every state pair and incurs quadratic pairwise work.

By Ibne Farabi Shihab, Joyanta Jyoti Mondal

rag reinforcement-learning benchmarks

arXiv AI

Jul 15

Learning When to Trust in Contextual Social Bandits

arXiv:2603. 13356v2 Announce Type: replace Abstract: Robust reinforcement learning typically assumes that feedback sources are either globally trustworthy or corrupted within a fixed global budget.

By Majid Ghasemi, Mark Crowley

reinforcement-learning safety

arXiv Machine Learning

Aug 11

The Cost of Adaptivity: Matching Lower Bounds Across Learning Problems

arXiv:2608. 08826v1 Announce Type: new Abstract: Adaptive procedures must work without nuisance information an oracle may use, such as a gradient scale or smoothness index, and robust procedures may have to answer queries whose coordinate and inspection time are chosen only after the data are seen.

By Ibne Farabi Shihab, Adria Binte Habib

benchmarks

arXiv Machine Learning

Aug 12

Detecting an Effect Is Not Learning to Act on It: A Reward-SNR Floor for LLM Acquisition Agents

arXiv:2608. 10441v1 Announce Type: new Abstract: Many pipelines can pay a per-example cost to acquire an auxiliary, model-derived observation -- an LLM's structured reasoning, a slow oracle, an expensive measurement -- and then must decide when the acquired signal is worth using.

By Ying Yuan

llms rag agents

Related stories

How Much of the Routing Gap Is Real? Decomposing the Router-to-Oracle Gap into Reproducible Specialist Advantage and Single-Draw Label Noise

Minimax PAC Bounds for Learning in Exogenous Contextual MDPs

Sub-Quadratic Bisimulation Metrics via Approximate Nearest Neighbors: Coverage-Augmented Guarantees and Computable Two-Sided Certificates

Learning When to Trust in Contextual Social Bandits

The Cost of Adaptivity: Matching Lower Bounds Across Learning Problems

Detecting an Effect Is Not Learning to Act on It: A Reward-SNR Floor for LLM Acquisition Agents