The paper introduces ARCBench, a configurable benchmark for measuring rule compliance in language‑model agents that compact long histories to fit context windows. Using ARCBench, the authors find that rule violations after compaction rarely stem from rule deletion; instead, agents often ignore the rule while it remains present in the compacted history. The study shows that the source of violations depends on how the rule is delivered and on the compaction strategy, with targeted compactor attacks significantly increasing violations, and that simply retaining a rule in the compacted summary does not guarantee compliance.
By Yiwei Zha, Qingyu Meng, Herbert Bos, Min Chen
arXiv:2606. 02965v2 Announce Type: replace Abstract: As large language models gain tool access and are deployed as autonomous agents capable of editing records, executing transactions, and modifying infrastructure, we still evaluate them based on the sole metric of task completion.
By Victor Ojewale, Suresh Venkatasubramanian
arXiv:2606. 22528v2 Announce Type: replace Abstract: Modern LLM agents increasingly rely on context compaction, summarization, or eviction to keep long-running sessions within a token budget.
By Shiyang Chen
arXiv:2606. 02965v1 Announce Type: new Abstract: Benchmarks for autonomous agents measure whether agents complete tasks, yet this framing is systematically blind to whether an agent should have proceeded at all.
By Victor Ojewale, Suresh Venkatasubramanian
arXiv:2607. 02599v1 Announce Type: cross Abstract: Tool-using LLM agents are usually evaluated by final-answer correctness or LLM judges.
By La\"ila Elkoussy (LRE, EPITA), Julien Perez (LRE)
arXiv:2608.22752v1 Announce Type: new
Abstract: A safety rule and an episodic log compete for the same tokens in an AI agent's context. When the budget overflows, both are summarized at the same rate...
By Saber Zerhoudi, Jelena Mitrovic, Michael Granitzer
arXiv:2608. 09828v1 Announce Type: cross Abstract: AI agents increasingly work inside systems that govern how they delegate tasks, move information, execute actions, and use shared resources.
By Abdullah X
arXiv:2608. 11392v1 Announce Type: cross Abstract: Long-running agents periodically compact their context, replacing the transcript with a model-generated summary.
By Ted Kwartler, Alan Aqrawi, Arian Abbasi
AgentAudit is an open, extensible framework that evaluates the full lifecycle of AI agents, assessing planning, tool selection, execution, memory, and reasoning across ten dimensions such as instruction integrity, security, and alignment. Unlike existing benchmarks that focus on single aspects, AgentAudit analyzes the entire execution trace to attribute failures to specific stages. The framework was tested on five large language models, revealing significant differences in trustworthiness even among models with similar task‑completion performance.
By Shrey Nag, Sachita, Abhishek Kumar Singh, Lipi Goel, Rajeshwar Singh Janwar
arXiv:2607. 26819v1 Announce Type: cross Abstract: Open source communities have been flooded with AI-generated contributions.
By Wenhao Yang, Runzhi He, Minghui Zhou
The paper investigates how context compaction—losing information to free up a language model’s context window—affects performance on long‑horizon agentic tasks. Using the OpenCode scaffold and five compaction strategies, the authors evaluate three frontier models on PaperBench, ProgramBench, and PostTrainBench, finding that compaction thresholds largely do not hurt performance for 1M‑context models, but can degrade GPT‑5.3‑Codex on certain benchmarks. The study also shows that compaction can silently drop critical task constraints, especially on ProgramBench and real user‑agent sessions from SWE‑chat, yet current benchmarks remain largely compaction‑friendly because objectives are fixed and progress is stored externally.
By Jeanne Salle, Hardik Bhatnagar, Matthias Bethge, Sahar Abdelnabi, Maksym Andriushchenko
arXiv:2510. 01359v2 Announce Type: replace-cross Abstract: Code-capable large language model (LLM) agents are embedded in software engineering workflows where they can read, write, and execute code, raising "jailbreak" stakes beyond text-only settings.
By Shoumik Saha, Jifan Chen, Sam Mayers, Sanjay Krishna Gouda, Zijian Wang, Varun Kumar