Skill-Use: Can LLMs Actually Use Skills in Agentic Harnesses?
Score
76首次从触发、合规、边界三维度刻画技能使用，增量显著。
基准构建严谨，任务和评分规则细致，评测过程可复现。
为智能体系统的技能调用模块提供了直接改进方向和评测工具。
正逢LLM智能体应用爆发期，技能使用问题是当前热点。
来自arXiv预印本，作者标注清楚，但尚未经同行评审。
Base 84.5 × T2 Source weight ×0.9 = 76
Curation line 70Selected
Why selected
The paper is the first to systematically decompose skill use and expose key bottlenecks, yet conclusions rely on limited models and harnesses, requiring broader validation.
AI commentary
「This benchmark turns 'whether models can actually use skills' into a measurable question, which is especially valuable for those building agent products and frameworks. The highlight is that skill use is not a fixed capability of the model but strongly depends on the harness, meaning we must design better invocation mechanisms, not just tune models. Recommended for every developer struggling with tool calling.」
Quick overview
Large language model (LLM) agents increasingly rely on skills—structured documents specifying when to act, which procedure to follow, and which tools are allowed. Existing evaluations mostly judge skill quality or task success, leaving unexamined whether an agent can recognize a relevant skill and apply it on its own. We introduce Skill-Use, a benchmark that evaluates skill use under progressive disclosure, where an agent sees only a skill's name and short description and must retrieve the full procedure before following it. Skill-Use separates three facets: Trigger, Compliance, and Boundary. A Skill-Use (SU) score combines the three and credits execution only after the skill is triggered. The benchmark pairs 79 real skills with 177 executable tasks across nine domains, runs in isolated Docker sandboxes, and scores via trajectory-based rubrics. Evaluating eight LLMs under two agent harnesses, we find reliable skill use remains out of reach: the strongest configuration reaches only 0.613, and both scores and rankings shift with the harness, indicating skill use is harness-conditioned rather than a fixed model property.
- Introduces the Skill-Use benchmark, evaluating skill use via Trigger, Compliance, and Boundary.
- Includes 79 real skills and 177 executable tasks across nine domains, run in isolated Docker sandboxes.
- Evaluates eight LLMs under two agent harnesses, with the best SU score only 0.613.
- Trigger and Compliance are independent bottlenecks, and model performance shifts with the harness.
Full text
This is an AI-generated Chinese digest; use "View original" for the source text.
Related items
FeaturedDeepSeek V4 Doubles Token Share on OpenRouter, Driven by Agentic Workloads
DeepSeek V4 在 OpenRouter 上的代币份额翻倍，智能体工作负载成主要驱动力
The rapid adoption of DeepSeek V4 signals a strong market preference for high-performance, low-cost inference models, especially as agentic use cases reshape AI deployment.
91
FeaturedEnabling Two Settings Triples ARC-AGI-3 Benchmark Scores
启用两项设置使ARC-AGI-3基准测试分数提升三倍
This finding is crucial! It shows that model performance depends not only on training but also on inference-time configuration. For AI engineers and researchers, this is a must-know optimization direction. Teams deploying models on resource-constrained devices in particular could find a low-cost way to boost capabilities.
91
FeaturedGPT-5.6: Fusing Frontier Intelligence with Frontier Efficiency
GPT-5.6：前沿智能与前沿效率的融合
This news is worth watching as it marks a shift from pure parameter scaling to efficiency competition. For AI developers and business decision-makers, GPT-5.6 means higher performance at lower cost, especially for long-context or agentic tasks previously hindered by inference expenses. OpenAI's move may redefine what 'frontier' means.
91
FeaturedLFM2.5-Encoders: New Models for Fast Long-Context Inference on CPU
LFM2.5-Encoder：面向CPU快速长上下文推理的新模型
This blog post catches my eye because it directly solves the pain point of running long-context LLMs on CPU. For teams or individual developers without high-end GPUs, the LFM2.5-Encoder series offers a ready-to-use open-source solution. Highly recommended for those working on RAG, document search, or code analysis—now a standard server can handle million-token documents.
91
FeaturedGPT-Red: OpenAI Automates Red Teaming via Self-Play to Boost Model Robustness
GPT-Red：OpenAI通过自我对弈实现AI红队自动化，提升模型鲁棒性
GPT-Red marks a milestone from human to automated red-teaming—essential for safety engineers and alignment researchers. While reducing cost, it also raises ethical questions about automated attacks, worth pondering for everyone in AI governance.
91
FeaturedOpenAI Launches GPT-5.5 Bio Bug Bounty Program
OpenAI推出GPT-5.5生物漏洞赏金计划
This program deserves attention as it pioneers the extension of AI bug bounty to biosecurity, offering a new research direction for the security community and a practical training ground for AI risk governance. It is particularly important for model safety researchers, biosecurity experts, and practitioners concerned about dual-use AI risks.
91
FeaturedvLLM Introduces Native-Speed Transformers Modeling Backend for Enhanced Inference Performance
vLLM推出原生速度Transformers建模后端，大幅提升推理性能
This update is critical for AI engineering teams: eliminating format conversion speeds up model iteration, and 2-5x speedup directly reduces GPU costs. If you're building LLM online services or batch inference, this is practically a must-read technical update.
90
FeaturedGoogle DeepMind Introduces SL2T, a Sign Language-to-Text Model for Deaf Users
Google DeepMind 发布手语转文本模型 SL2T，助力聋人用户
This news matters not only as a technical breakthrough but also as a sign that AI is beginning to address the needs of marginalized groups. For developers, the architecture and technical approach of SL2T merit deep study; for accessibility advocates, it's a key case for advancing industry standards.