arXiv:2606. 15579v1 Announce Type: new Abstract: We propose Base Sequence Analysis, a framework that encodes the runtime behavior of LLM-powered autonomous agents into compact symbolic sequences using a four-letter alphabet: X (Explore), E (Execute), P (Plan), and V (Verify).
arXiv:2606. 07682v1 Announce Type: cross Abstract: AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex environments.
BekchiAI introduces a benchmark and platform for evaluating large language model agents. The benchmark comprises 13 tool‑using ReAct agents across seven task categories, totaling 2,057 deterministic test tasks with verifier‑checkable gold answers. The platform offers web‑based observability, token and latency telemetry, and remote run termination for live agents.
arXiv:2605. 12925v3 Announce Type: replace-cross Abstract: Evaluation of software engineering (SWE) agents is dominated by a binary signal: whether the final patch passes the tests.
AgentServeSim is a simulation framework designed to model the execution of large language model (LLM) agent programs, capturing cross‑turn key‑value (KV) state retention, successor turn release, and scheduling decisions. Unlike existing simulators that operate on request streams, AgentServeSim treats the entire agent program as a single unit of execution, using a Program Control Block, Program Orchestrator, Retention Plane, and Dispatch Plane to emulate realistic serving dynamics. Validation against real vLLM deployments on two GPU platforms shows mean job completion time errors below 5.5%, and the simulator enables automated policy search that improves mean JCT by up to 2.8% over hand‑written policies. whyItMatters":"The simulator provides a realistic, CPU‑based tool for evaluating and optimizing LLM agent serving policies, achieving high fidelity to real deployments and enabling measurable performance gains."
arXiv:2607. 10059v1 Announce Type: new Abstract: Agent systems based on large language models (LLMs) are increasingly deployed for autonomous tasks, yet existing evaluations mostly focus on task success rather than whether agents know when to abstain.
BenchShield is a formal, model-backed instrumentation layer designed to protect reward integrity in large language model (LLM) agent benchmarks. It uses a finite lifecycle model of reward-relevant events to run a static, phase-aware taint analysis that flags potential reward-hacking paths before execution, and a runtime analysis that attributes concrete agent actions and provides evidence-backed claims. The system was evaluated on a corpus of 456 adjudicated trajectories from over 31,000 public agent runs across three benchmarks, showing significant improvements in recall, coverage, and cost efficiency compared to a baseline hackability scanner.
Mid‑Harness proposes a test‑time compute strategy that samples and verifies candidate actions before execution, keeping the underlying generator and harness unchanged. Experiments show that with a strong verifier, sampling more actions significantly boosts success rates—e.g., a GPT‑5.6 verifier raises Pass@1 from 50.00 % to 68.03 % on TerminalBench‑Lite using eight samples. The approach also improves performance across various models, benchmarks, and harnesses, demonstrating that action scaling is a promising target for enhancing terminal agent reliability.
The paper introduces VHD-Play, a pipeline that first samples and solves a mathematical model before generating agentic reinforcement learning environments, ensuring that dynamics and evaluation are aligned from the outset. This approach yields 3,300 diverse environments at a low cost and significantly improves the performance of a large language‑model agent (Qwen3.6‑35B‑A3B) across multiple diagnostic families and external benchmarks. The study demonstrates that stateful interaction is a key factor in learning gains and that scaling the training substrate can further enhance performance.
The paper introduces RuVerBench, a benchmark with 2,458 instances for evaluating the reliability of Large Language Models acting as judges (LaaJ) in verifying rubric compliance within agentic scenarios such as deep research and agentic coding. It systematically meta‑evaluates frontier LLMs, revealing that even the most advanced models perform well yet still produce substantial noise. The study also examines how prompt design, batching, and majority voting affect verification accuracy, noting that weaker models are more prompt‑sensitive, batched verification trades accuracy for efficiency, and majority voting offers diminishing returns.
arXiv:2606. 17546v1 Announce Type: new Abstract: Self-evolving LLM-based agents improve mainly by changing their agent harness: the structured execution layer around a base model, including prompts, memory, tools, middleware, runtime state, and the model-tool interaction loop.
arXiv:2609.06972v1 Announce Type: cross Abstract: LLM agents complete tasks by issuing sequences of tool calls, and every observation they read is a channel through which an indirect prompt injection...
LatentSift is a token‑free, execution‑free filtering method for software engineering agents that replaces the initial LLM‑based verifier. It represents each candidate trajectory using the policy’s hidden states—reasoning, observation, and function‑call states—and compares them against banks of successful and unsuccessful states collected during training. By fusing distance scores with a learned linear score, LatentSift retains promising candidates, reducing EF‑verifier tokens by 66.6–81.0% and total verification tokens by 49.1–62.1% while maintaining or improving performance on DeepSWE‑Preview.