Large language model (LLM) agents are increasingly expected to assist users in completing tasks. However, existing benchmarks provide limited support for evaluating whether agents can carry out office-suite workflows at a reasonable cost.
arXiv:2607. 27155v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly expected to assist users in completing tasks.
arXiv:2609.06059v1 Announce Type: new Abstract: As large language models evolve from question-answering systems into general-purpose agents, evaluation must move beyond static answer correctness to a...
arXiv:2608. 00101v1 Announce Type: cross Abstract: AI coding agents like GitHub Copilot, Claude Code, and Codex interleave multi-step LLM inference with tool execution, creating a workload different from chatbots.
arXiv:2604. 13072v2 Announce Type: replace-cross Abstract: OpenClaw-style personal assistants extend LLM agents from isolated tool use to open-ended, stateful, and personalized software environments.
arXiv:2606. 30560v1 Announce Type: cross Abstract: Coding agents are rapidly becoming a major application of agentic LLMs, but serving them efficiently remains challenging.
arXiv:2604. 08523v2 Announce Type: replace-cross Abstract: AI agents may be able to assist with emails and documents, but can they reliably complete everyday online workflows on real websites?
arXiv:2603. 14501v2 Announce Type: replace-cross Abstract: Large Language Models excel in high-resource programming languages but struggle with low-resource ones.
UniACE is a unified framework that standardizes the evaluation of large language model (LLM) agents by representing each benchmark as an instruction–tool–environment triplet and running models through a shared, task‑agnostic harness in isolated runtimes. It preserves native success criteria, offers an offline mode for dynamic‑resource tasks, and standardizes efficiency metrics, execution records, and failure attribution. Applying UniACE to 7 benchmarks across 24 domains and 15 models revealed significant score shifts, ranking reversals, and sensitivity to evidence representation, highlighting the impact of evaluation configuration on reported agent performance.
arXiv:2511. 02734v3 Announce Type: replace Abstract: Current evaluations of Large Language Model (LLM) agents primarily emphasize task completion, often overlooking resource efficiency and adaptability.
OS-Marathon is a new benchmark that tests computer‑use agents on vast‑horizon, repetitive tasks, covering 100 tasks across five scenarios and ten domains. The study shows that current state‑of‑the‑art agents perform poorly on these tasks, and that simply decomposing workflows into subtasks does not solve the problem. Introducing a cost‑friendly personalization method called GraphDemo, which adapts agents from a single human demonstration, improves performance, highlighting the value of human guidance for these challenging tasks.
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.
arXiv:2511. 00802v2 Announce Type: replace-cross Abstract: With data-driven development now widely adopted, online A/B testing is an established method for measuring the effects of new technologies.