arXiv:2606. 20363v1 Announce Type: new Abstract: Explicit skill libraries make computer-using agents easier to inspect, but it remains unclear whether such libraries can be mined from interaction data in a way that improves downstream policies.
arXiv:2605. 09192v2 Announce Type: replace Abstract: Agent skills can remarkably improve task success rates by using human-written procedural documents, but their quality is difficult to assess without environment-grounded verification.
UniToolCall introduces a unified framework for tool-use in large language model agents, standardizing toolset construction, dataset generation, and evaluation. The framework aggregates over 22,000 tools and creates a hybrid training corpus of more than 390,000 instances by combining ten public datasets with synthetically generated, structurally controlled trajectories. It models diverse interaction patterns—single‑hop vs. multi‑hop, single‑turn vs. multi‑turn, serial vs. parallel execution—and adds an Anchor Linkage mechanism to enforce cross‑turn dependencies, while converting seven public benchmarks into a common Query–Action–Observation–Answer format for fine‑grained evaluation.
arXiv:2610.33907v1 Announce Type: new Abstract: Computer-using agents are increasingly capable of performing complex tasks in the digital world. However, high-quality interaction trajectories remain...
arXiv:2609.05837v1 Announce Type: new Abstract: LLM-based agents are increasingly deployed in real-world applications through tool-use APIs, yet training them for specific environments remains fundam...
AgentHorizon is a benchmark comprising 1,373 computer‑use tasks with instruction‑trajectory pairs collected from 166 hours of human‑recorded activity across three operating systems. The benchmark uses a paired design that swaps instructions to create negative tasks, testing judges’ ability to distinguish successful trajectories from those that completed a similar but incompatible request. Eleven judges were evaluated, with GPT‑5.5 achieving the highest balanced accuracy of 80.9% on the full trajectory subset, while tool‑use benefits some models but harms others, revealing large variability in judges’ performance.
LiteGUI is a new framework for training lightweight GUI agents that tackles the long‑horizon, multi‑path nature of GUI interactions. It introduces a two‑stage post‑training pipeline: Guided On‑Policy Distillation, which uses human‑verified multi‑solution action annotations to provide privileged teacher guidance at each state, and Multi‑Solution Dual‑Level GRPO, which combines action‑level supervision with history‑conditioned planning‑quality supervision while handling multiple valid actions. The authors also present a scalable data generation pipeline and the Lite‑Dataset, enabling extensive evaluation across GUI benchmarks and demonstrating substantial performance gains over existing training paradigms.
arXiv:2607. 21596v2 Announce Type: replace Abstract: Large language model agents can adapt to complex tasks by constructing workflows at inference time, but procedures discovered in one episode are usually discarded after execution.
The paper proposes a two-stage adaptive training framework for GUI agents that addresses both where to learn and how far to roll out during training. It creates reusable intermediate checkpoints from expert trajectories and selects informative pivots using action disagreement to focus one-step reinforcement learning on uncertain decisions. The method then progressively expands policy responsibility through adaptive Local, Suffix, and Root training, enabling repeated training from intermediate states and reducing environment interaction by 56‑61% while improving task performance on MobileGym and CUA‑Gym‑Hub.
Computer-using agents can perceive rich software interfaces, yet their decisions often lack visual procedural memory: they may recognize individual controls without identifying which familiar workflow is active, which control matters next, or what evidence would confirm progress. Raw interaction traces preserve such information but are long and noisy to condition on, whereas text-only skills often omit the visual state that makes a procedure applicable.
SkillGym is an automatic pipeline that generates verifiable environments for training skill-use agents. It crawls internet skills, filters for reproducible workflows, and uses a builder‑reviewer process to create difficulty‑controlled tasks with reference solutions and verifiers. The system builds 6.8k environments, collects 19k successful trajectories, and fine‑tunes LLMs from 2B to 122B parameters, improving performance and skill invocation rates.
arXiv:2607. 25853v1 Announce Type: new Abstract: Skills have become an important abstraction for enabling large language model (LLM) agents to reuse past experience in long-horizon interactive tasks.
TRACER is a retrieval front‑end that indexes tool‑call sequences in a trie and uses a learned blend of empirical and contextual priors to rank tool continuations. In a mixed retail‑and‑telecom environment, it boosts GPT‑5.4’s tool‑selection accuracy by 4.7 points while cutting input tokens per step by 80%, exposing only 5 of 128 tools. Across six environments and four models, TRACER improves tool‑selection accuracy in 21 of 24 model‑environment pairs, with average gains of 6 to 10 points, and matches full‑registry success with 61% fewer tokens per step in autonomous execution.