Mingbird is a local‑first agent harness designed for small open‑weight language models (2–9 B) that run on ordinary laptops. It introduces ten mechanisms—such as a byte‑level net‑zero prefill budget, a finish gate that re‑reads the task before accepting completion, and signature‑level loop detection—to address common failure modes that arise from the harness rather than the model itself. In controlled experiments on the LRAB benchmark and the $ au^2$‑bench, Mingbird achieves higher overall scores (0.886 and 0.856 respectively) compared to other harnesses, and its ablation studies show that each mechanism contributes measurable performance gains.
By Hao Wang, Ting Huang
arXiv:2607. 28074v1 Announce Type: cross Abstract: Computer-use agents learn from what their actions change, so training one needs applications it can act on, break and reset.
By Yash Pandya, Sahil Gupta, Sarthak Harne, Archana Yadav, Kavyansh Chourasia, Hussein Mozannar, Vibhav Vineet, Sara Abdali, Corby Rosset, Yash Lara, Ahmed Awadallah, Ece Kamar, Akshay Nambi
arXiv:2606. 24311v1 Announce Type: new Abstract: As large language model (LLM) agents are applied to longer tasks, they increasingly modify workspace state across multiple rounds of iteration.
By Kailong Ren, Fubo Sun, Jiachen Liu, Liu Yang, Zimo Yin, Jiaying Li, Congli Yin, Ming He, Yu Huo, Jiawei Liu, Zeping Chen, Yubin Huangfu, Ronghua Li, Yixuan Wu, Xing Su, Yanzhi Xu, Likang Wu, Hongke Zhao, Lei Zhang, Xiaohui Geng, Jianping Fan
arXiv:2603.01209v3 Announce Type: replace
Abstract: In CodeAct, language-model agents write Python that calls tools and use execution feedback to choose actions. Persistent runtimes preserve Python v...
By Victor May, Van Khue Nguyen, Aaditya Salgarkar, Yishan Wang, Diganta Misra, Huu Nguyen
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.
By Pengyu Zhu, Lijun Li, Yaxing Lyu, Qianxin Luo, Jingyi Yang, Yi Liu, Tingfeng Hui, Xinyu Yuan, Li Sun, Sen Su, Jing Shao
arXiv:2608. 16381v1 Announce Type: new Abstract: Agentic systems often organize execution and state around a single conversation, model invocation, or agent instance, even when real work spans many calls and stages.
By Zhenhang Nie (iFLYTEK Co., Ltd., Hefei, China), Gui Zheng (iFLYTEK Co., Ltd., Hefei, China), Xudong Sun (iFLYTEK Co., Ltd., Hefei, China), Tailong Zhu (iFLYTEK Co., Ltd., Hefei, China), Bin Zhang (iFLYTEK Co., Ltd., Hefei, China)
Computer-use agents learn from what their actions change, so training one needs applications it can act on, break and reset. The applications that matter most are login-gated and stateful, so synthetic environments stand in for them.
arXiv:2608.22510v1 Announce Type: new
Abstract: Agent benchmarks often evaluate only final answers even when agents run on stateful runtimes. We argue this under-specifies what is being evaluated: th...
By YuanHang Xiao
arXiv:2609.24165v1 Announce Type: new
Abstract: Synchrotron data reduction, detector calibration followed by azimuthal integration of terabyte-scale diffraction series, is a multi-step, expert-bound...
By Pawan K. Tripathi, Hemant Sharma, Andrew Chuang, Mathew J. Cherukara
arXiv:2606. 29537v2 Announce Type: replace Abstract: Existing computer-use benchmarks fail to capture the realism, complexity, and long-horizon demands of real-world computer use, limiting their ability to reveal the limitations of frontier agents.
By Mengqi Yuan, Zilong Zhou, Xinzhuang Xiong, Weiming Wu, Jiayang Sun, Jiamin Song, Kaiqian Cui, Bowen Wang, Haoyuan Wu, Yitong Li, Dunjie Lu, Haikong Lu, Qi Zhen, Xinyuan Wang, Jiaqi Deng, Yuhao Yang, Cheng Chen, Boyuan Zheng, Alex Su, Xiao Yu, Hao Zou, Saaket Agashe, Xing Han Lu, Manpreet Kaur, Zhengyang Qi, Vincent Sunn Chen, Frederic Sala, Dayiheng Liu, Junyang Lin, Zhou Yu, Yu Su, Siva Reddy, Xin Eric Wang, Peng Qi, Tianbao Xie, Tao Yu
The paper introduces a coroutine-bridge harness that lets a language model emit a Python program to manage tool calls in the CAR-bench evaluation. By decoupling model invocations from tool round-trips, the approach reduces model calls to a median of two per task while maintaining seven agent turns, achieving a median latency of 1.8 s on a Cerebras gpt‑oss‑120b. The harness achieved 60.0 % Pass³ on the official hidden evaluation, outperforming the baseline by 4.5× and matching frontier-model agents on GPT‑5.5, all while keeping the prompt largely cached and minimizing input compute.
By Ivan Matveev
arXiv:2606. 29537v1 Announce Type: new Abstract: Existing computer-use benchmarks fail to capture the realism, complexity, and long-horizon demands of real-world computer use, limiting their ability to reveal the limitations of frontier agents.
By Mengqi Yuan, Zilong Zhou, Xinzhuang Xiong, Weiming Wu, Jiayang Sun, Jiamin Song, Kaiqian Cui, Bowen Wang, Haoyuan Wu, Yitong Li, Dunjie Lu, Haikong Lu, Qi Zhen, Xinyuan Wang, Jiaqi Deng, Yuhao Yang, Cheng Chen, Boyuan Zheng, Alex Su, Xiao Yu, Hao Zou, Saaket Agashe, Xing Han Lu, Manpreet Kaur, Zhengyang Qi, Vincent Sunn Chen, Frederic Sala, Dayiheng Liu, Junyang Lin, Zhou Yu, Yu Su, Siva Reddy, Xin Eric Wang, Peng Qi, Tianbao Xie, Tao Yu