论文精选 82arXiv

Rethinking Self-Evolving Agents: Do We Still Need Prescribed Optimization Pipelines?· 重新思考自进化代理

Self-evolving agents are usually built around prescribed optimization pipelines: the framework decides how to gather evidence, revise a persistent artifact, select candidates, and stop. We ask whether this task-specific procedure remains necessary when a frontier model acts as the optimizer. We introduce Open-Ended Optimization (OEO), which keeps the objective, permitted interactions, resource budget, data boundary, and evaluation fixed while allowing the optimizer to compose the improvement process online. We compare OEO with two complementary prescribed approaches: SkillOpt, a staged pipeline with bounded edits, and GEPA, a reflective evolutionary search. Across 14 head-to-head comparisons over 8 benchmark-target-model settings, GPT-5.5-driven OEO records 12 wins, 1 tie, and 1 narrow los

领域：cs.AI作者：Hui Xue、Fan Yang

#cs.AI

📎 发布于 08-10 22:10🔗 阅读原文（arxiv.org）📄 PDF

A Deep Generative Model for Synthesizing Labeled Wireless Signals· 用于合成带标签无线信号的深度生成模型

Wireless signals with position-related labels are pivotal for both performance evaluation …

领域：cs.AI作者：Yuxiao Li、Keke Hu、Santiago Mazuelas

📎 arXiv🕒 09-05 01:44🔗 arxiv.org

论文精选 60已解读

Multi-Step Tool-Calling over Korean Open Public APIs: A Benchmark and a Data-Synthesis Recipe· 韩国开放公共API的多步调用基准

Data-sovereignty regulations increasingly require public institutions to deploy open-sourc…

领域：cs.AI作者：Dain Kim、Eungi Cho、Kyumin Kim

📎 arXiv🕒 09-05 01:44🔗 arxiv.org

论文精选 60已解读

Necessary or Sufficient? Evaluating LLM Explanations With Behavioural Evidence· 必要还是充分？用行为证据评估LLM解释

LLM decision components that can operate within agent workflows often produce action-relev…

领域：cs.AI作者：Urja Pawar、Rajitha Ramanayake、Nabeel Kemal

📎 arXiv🕒 09-05 01:37🔗 arxiv.org

论文精选 60已解读

Molecular Déjà Vu: Digit-Level Retrieval of Published Values in Frontier Language Models· 分子属性预测中的大模型文字级检索

Large language models (LLMs) are increasingly evaluated on molecular property benchmarks, …

领域：cs.AI作者：Matthias Busch、Marius Tacke、Sviatlana V. Lamaka

📎 arXiv🕒 09-05 01:32🔗 arxiv.org

论文精选 60已解读

CUA-Universe: A Scalable and Dynamic Environment for Hybrid GUI+CLI Agents· CUA-Universe: GUI+CLI 混合代理的可扩展动态环境

Computer-use agents have advanced on benchmarks like OSWorld and AndroidWorld, but still a…

领域：cs.AI作者：Haoting Shi、Wenhao Wang、Weicheng Fang

📎 arXiv🕒 09-05 01:26🔗 arxiv.org

本站内容由 LLM 精选聚合，原文版权归 arXiv 所有 · 摘录仅供参考