This work formalizes four representative agent payment protocols: x402, MPP, ACP, and AP2 in Tamarin, and constructs source-grounded models that capture each protocol's roles, state, trust assumptions, and lifecycle transitions.
Ke Jiang, Mo-Han Yu, Yuan-Yi-Chun-Min-Chieh Chang et al.· 0 citations
The theory that explains the blindness, the theory that explains the blindness, the two-population conservation law, under which every in-loop metric improves while true quality does not, and a synthetic study that validates the mechanism are given.
D. Rajput, Nirdesh Chauhan, S.Rao Kosaraju· 0 citations
3R-Bench (Refusal, Repetition, and Revision), a benchmark of 150 real-world cybersecurity requests augmented with two adversarial conversational settings, is introduced and eight LLMs are evaluated, finding that prior assistant behavior strongly changes responses to an unchanged request.
Rui Yang, Yang Hong, Yi-Chao Xu et al.· 0 citations
OpenAgentFlow is presented, a control-plane/action-plane architecture that establishes the action-commit boundary as a shared enforcement interface and shows that a shared action-commit boundary provides a practical basis for system-wide governance across heterogeneous agent execution paths.
Dong-Sheng Chen, Xiang-Yu Zhao, Xin Yao et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Gpt-oss-120b, a mixture-of-experts model with only $\sim$5.5B active parameters per token, carries the full state across all calls and returns the correct digest on a majority of completed runs, and localizes the residual failures by origin, separating state-carrying from arithmetic and from serving.
Experiments show that Knowledge Trap reduces surrogate Agreement by 6.2\% on average without degrading legitimate-user accuracy, outperforming existing defenses that impose measurable user impact, and suggest that defending knowledge-space traversal is a practical direction for mitigating LLM extraction attacks.
Overall, the results show that SNN design choice, evaluation protocol, spike-timing representation, and computational activity should be examined separately when applying SNNs to static network-flow data.
Raj Patel, David Amebley, Taye Akinrele et al.· 0 citations
This work presents TrustFlip, a novel attack that weaponizes consistency-based defenses to poison the trust assigned to benign vehicles, and introduces TrustReflect, a lightweight self-reflection mechanism that marks disputed regions as uncertain and excludes them from trust evaluation, reducing the attack success rate...
Yu-Tong Liu, Chenyi Wang, Ming F. Li et al.· arXiv.org· 0 citations
It is argued that evaluation should move beyond static question answering and model accuracy towards workflow-level assessment of evidence quality, tool use, policy and invariant compliance, staged execution, recovery, calibration, cost, and human intervention.
Muhammad Bilal, Jon Crowcroft, Rui-Zhi Wang et al.· arXiv.org· 3 citations· ⚡1
This work introduces a deterministic full-chain memorization mechanism that locks onto token-level secrets in dynamic computation flows via online tensor-rule matching, and leverages value-gradient decoupling to stealthily inject attack gradients, overcoming gradient drowning to force model memorization.
Zi Li, Tianyang Zhou, Wenze Li et al.· arXiv.org· 0 citations
AgenTRIM is introduced, a framework for detecting and mitigating tool-driven agency risks without altering an agent's internal reasoning that provides a practical, capability-preserving approach to safer tool use in LLM-based agents.
Roy Betser, Amit Giloni, Shamik Bose et al.· arXiv.org· 14 citations· ⚡2
JAWS-BENCH(Jailbreaks Across WorkSpaces), a benchmark spanning three escalating workspace regimes mirroring attacker capability, is presented, indicating that JAWS-BENCH can be reused across multiple agent frameworks.
Shoumik Saha, Jifan Chen, Sam Mayers et al.· Transactions of the Associat...· 8 citations· ⚡2