aiqbench analysis
This paper introduces the Counterfactual Memory Audit (CMA), a rigorous protocol for evaluating whether robot policies genuinely rely on historical context or merely react to current inputs. By testing frozen policies across reconverged histories, the study reveals significant gaps between memory sensitivity and reliable decision-making in physical robotic systems. These findings are critical for advancing trustworthy embodied AI by establishing standards for verifying that past experiences actually guide future actions.
Memory That Changes Action Is Not Memory That Guides It: Counterfactual Auditing of History-Conditioned Robot Policies
arXiv:2609.27247v1 Announce Type: new Abstract: A robot returning a block to its origin tray may encounter two task-consistent pasts that reconverge to the same current input but warrant different actions. Yet memory-policy evaluations often rely on task success or action change under memory perturbation, neither of which establishes that memory guides the decision. We propose the \textbf{Counterfactual Memory Audit (CMA)}, an evaluation protocol that crosses two histories at a verified-identical present, queries a frozen policy under common randomness, and evaluates each saved action under both pasts. This separates memory sensitivity, warranted choice, matched-world physical value, and per-pair reliability. On Mem-0, every audited Put Back pair changes action, but only $20/64$ pairs are fully reliable; at a later Swap decision, all paired actions change while both memories select the same branch. Native interventions further show closed-loop influence: replacing the history bank redirects behavior toward the replaced content, while restoring a 4096-byte protected anchor recovers $38.9$ points of Swap success lost to injected bank faults. On a dual-arm physical platform, memory changes saved actions, yet five of nine completed Put Back manipulations reach the wrong target. These results show that a robot can remember and react without reliably using memory to choose the behavior its past warrants. CMA provides a decision-level audit for distinguishing these cases.
Abstract:A robot returning a block to its origin tray may encounter two task-consistent pasts that reconverge to the same current input but warrant different actions. Yet memory-policy evaluations often rely on task success or action change under memory perturbation, neither of which establishes that memory guides the decision. We propose the \textbf{Counterfactual Memory Audit (CMA)}, an evaluation protocol that crosses two histories at a verified-identical present, queries a frozen policy under common randomness, and evaluates each saved action under both pasts. This separates memory sensitivity, warranted choice, matched-world physical value, and per-pair reliability. On Mem-0, every audited Put Back pair changes action, but only $20/64$ pairs are fully reliable; at a later Swap decision, all paired actions change while both memories select the same branch. Native interventions further show closed-loop influence: replacing the history bank redirects behavior toward the replaced content, while restoring a 4096-byte protected anchor recovers $38.9$ points of Swap success lost to injected bank faults. On a dual-arm physical platform, memory changes saved actions, yet five of nine completed Put Back manipulations reach the wrong target. These results show that a robot can remember and react without reliably using memory to choose the behavior its past warrants. CMA provides a decision-level audit for distinguishing these cases.
Robotics (cs.RO)
arXiv:2609.27247 [cs.RO]
(or arXiv:2609.27247v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.27247 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 302 words
Related coverage

Deconfusing Corrigibility
AI Alignment Forum · Oct 17, 2026

Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026

Six Principles of Voice Design: Designing for the Ear and Encoding It into Agent Script
Salesforce Blog · Oct 8, 2026
Performance Profiling on AMD GPUs – Part 5: Profiling-Driven Kernel Optimization with an AI Code-Assist Tool
AMD · Jul 16, 2026