The paper introduces Pufibara, an agent harness designed to maintain engineering state and evidence across revisions in Modelica-based physical system modeling. It also presents a 232-task Modelica Agent Workflow Benchmark covering model repair, generation, and tuning, evaluated by an external benchmark-owned evaluator. Experiments show Pufibara outperforms Claude Code in task success and resource efficiency across two LLM backends.
By Zizhe Wang
arXiv:2605. 27898v2 Announce Type: replace Abstract: As LLMs are increasingly deployed as agents, reliable assessment of their agentic capabilities has become essential.
By Pengyu Zhu, Lijun Li, Yaxing Lyu, Qianxin Luo, Jingyi Yang, Yi Liu, Tingfeng Hui, Xinyu Yuan, Li Sun, Sen Su, Jing Shao
The paper introduces ReviveBench, a benchmark designed to evaluate coding agents’ ability to revive non‑running software and reconstruct industrial engines from open specifications. It comprises two families of tasks—revival (ten tasks addressing dependency issues, missing modules, legacy builds, and GPU models) and reconstruction (thirteen tasks covering numerical, geometric, hardware, and transactional systems). The benchmark uses hidden verifiers calibrated against native environments, engineering tools, or reference implementations, and the authors report that the strongest evaluated model passes all revival tasks and most reconstruction tasks, while also uncovering verifier defects that highlight measurement error in executable verification.
By Tianyu Liu, Dingyuan Dai, Yufan Du, Zhen Yang
Computer-use agents are increasingly evaluated in realistic desktop environments, but existing benchmarks provide limited coverage of professional engineering workflows whose outputs are persistent, s...
EngiAI introduces a capability-based evaluation framework for tool-connected engineering agents, assessing workflow execution, retrieval-assisted parameter selection, HPC orchestration, and training-code authoring using execution traces and engineering artifacts. The framework was applied to four LLM backends on EngiBench Beams2D and Photonics2D, revealing that proprietary models outperform open-source ones in workflow completion and HPC orchestration, while indexed retrieval improves parameter selection. The study demonstrates that evaluating distinct skills separately provides clearer insight into failure mechanisms than end-to-end success rates alone.
By Gioele Molinari, Florian Felten, Soheyl Massoudi, Mark Fuge
Harness-Bench is a diagnostic benchmark that evaluates how different harness configurations affect LLM agent performance in realistic workflows. It tests 106 sandboxed offline tasks across multiple model backends, recording artifacts, execution traces, and usage statistics to analyze completion, process quality, efficiency, and failure behavior. The study finds significant variation in outcomes across model-harness pairings, indicating that agent capability should be reported at the configuration level rather than solely by the base model.
By Yilun Yao, Xinyu Tan, Chao-Hsuan Liu, Yaoming Li, Zhengyang Wang, Zongwei Lv, Wenhan Yu, Zhewen Tan, Yuxuan Tian, Lin Sun, Xiangzheng Zhang, Guangxiang Zhao, Tong Yang
arXiv:2607. 02436v1 Announce Type: cross Abstract: Agentic coding assistants are increasingly given extra capabilities, such as browser based testing tools and design oriented system prompts, on the assumption that more capability yields better software.
By Achint Mehta
EngiWorld is a new benchmark that tests autonomous agents across the full engineering design loop, covering 1,301 expert‑curated tasks in six domains (CAD, CAE, CAM, BIM, EDA, and 3D visualization) and 26 professional software platforms with both GUI and CLI interfaces. It introduces an artifact‑centric evaluation method that programmatically verifies geometric validity, physical feasibility, and rule compliance of both final and intermediate artifacts, scoring tasks continuously rather than with binary success. Initial tests of seven frontier models show a large capability gap, with the best model scoring only 44.3 on the EngiScore and just 3.6% of multi‑software attempts succeeding.
By Hongcheng Gao, Hailong Qu, Yu Lei, Henghui Sun, Haoyang Li, Yipeng Wei, Naihao Xue, Xiaohan Yu, Zhuo Tao, Yihe Zang, Yajiao Wang, Jingyi Tang, Yi Li, Jingjing Zhou, Jie Luo, Bohan Zeng, Chengyu Shen, Hao Jiang, Chong Chen, Bowen Qu, Olive Huang, Zeqiang Wang
The paper introduces a framework for evaluating large‑language‑model agents by focusing on the interaction between the model and its execution harness. It defines five complementary metrics—Valid Completion Rate, Model‑Harness Fit, Cross‑Model Fit, Interaction Residuals, and Token Load—to assess how reliably agents finish tasks, how well harnesses support specific models, and the overall suitability across models. Using these metrics, the authors evaluate four general‑purpose harnesses and two native systems on six benchmarks, finding that harness rankings differ by model and task and that some harnesses consistently achieve high valid completion rates.
By Shao Zhang, Hangfan Zhang, Yuanyuan Shi, Hao Li, Min-Le Su, Kangcong Li, Jingping Wang, Chunjiang Mu, Chen Zhang, Yang Chen, Bo Zhang, Shuyue Hu
The paper introduces TraceTrial-Bench, a benchmark that evaluates the reliability of AI agent harnesses across seven components—context, execution, governance, lifecycle, observability, tools, and verification—using 95 expert-level tasks with executable or state-based ground truth. It employs a Trajectory Tracing module to record agent execution, a two-layer evaluation framework to separately assess agent performance and harness reliability, and a Multi-source Evidence Acquisition module to gather evidence for LLM-based evaluation. The benchmark’s analysis of 3,420 runs across four harnesses and nine models shows harness scores between 59.8% and 72.0%, with 31.1% of runs succeeding despite failing at least one harness criterion, highlighting the need for auditable and reproducible harness evaluation.
By Zhiyong Cao, Hengjie Zhu, Yongxin Wang, Xuchen Li, Jintao Li, Shuxia Liu, Dunqiang Liu, Bo Wang, Kairan Zhang, Cong Lin, JiangDi Yu, Siyuan Liu, Sijia Yao, Yi-Liang Wei, Lingxu Ma, Aijia Wang, Anxin Chen, Zixian Wei, Xiang Ruan, Ruizhi Xu, Ruqian Shi, Baihua Xiao, Sicheng Zhou
arXiv:2607. 17686v1 Announce Type: cross Abstract: Modern software teams have mature tools for low-level testing, such as pytest, JUnit, and Jest, which make it inexpensive to write unit tests and run them on every commit.
By Mansur Arief, Nur Ahmad Khatim, Ali Akarma, Ahmad Alfan Alfian Irfan
arXiv:2609.16251v1 Announce Type: new
Abstract: Computer-use agents are increasingly evaluated in realistic desktop environments, but existing benchmarks provide limited coverage of professional engi...
By Zihan Dong, Yuanzhe Liu, Zhiyuan Ma, Qishi Zhan, Dehan Kong, Guohao Li, Kaixin Li