自动驾驶

Evaluation of Neural Networks for Image Recognition Applications: Designing a 0-1 MILP Model of a CNN to create adversarials

Lucas Schelkes

专题命中 仿真评测 ：self-driving(abstract)；分类 cs.CV

Comments Thesis Bergische Universität Wuppertal

An Auto-tuning Framework for Autonomous Vehicles

Haoyang Fan, Zhongpu Xia, Changchun Liu, Yaqin Chen, Qi Kong

专题命中 仿真评测 ：autonomous driving(abstract)；分类 cs.RO

Comments 7 pages, 9 figures, 2 tables

Understanding Humans in Crowded Scenes: Deep Nested Adversarial Learning and A New Benchmark for Multi-Human Parsing

Jian Zhao, Jianshu Li, Yu Cheng, Li Zhou, Terence Sim, Shuicheng Yan, Jiashi Feng

专题命中 仿真评测 ：autonomous driving(abstract)；分类 cs.CV

Comments The first three authors are with equal contributions

What Face and Body Shapes Can Tell About Height

Semih Günel, Helge Rhodin, Pascal Fua

专题命中 仿真评测 ：autonomous driving(abstract)；分类 cs.CV

Sim4CV: A Photo-Realistic Simulator for Computer Vision Applications

Matthias Müller, Vincent Casser, Jean Lahoud, Neil Smith, Bernard Ghanem

专题命中 仿真评测 ：autonomous driving(abstract)；分类 cs.CV

Comments Published at the International Journal of Computer Vision (IJCV), 2018

3D Registration of Aerial and Ground Robots for Disaster Response: An Evaluation of Features, Descriptors, and Transformation Estimation

Abel Gawel, Renaud Dubé, Hartmut Surmann, Juan Nieto, Roland Siegwart, Cesar Cadena

专题命中 仿真评测 ：LiDAR(abstract)；分类 cs.RO

Comments Awarded Best Paper at the 15th IEEE International Symposium on Safety, Security, and Rescue Robotics 2017 (SSRR 2017)

Human Pose Forecasting via Deep Markov Models

Sam Toyer, Anoop Cherian, Tengda Han, Stephen Gould

专题命中 仿真评测 ：autonomous driving(abstract)；分类 cs.CV

Comments Accepted to DICTA'17

Momo: Monocular Motion Estimation on Manifolds

Johannes Graeter, Tobias Strauss, Martin Lauer

专题命中 仿真评测 ：autonomous driving(abstract)；分类 cs.CV

Maximum Resilience of Artificial Neural Networks

Chih-Hong Cheng, Georg Nührenberg, Harald Ruess

专题命中 仿真评测 ：self-driving(abstract)；分类 cs.AI

Comments Timestamp research work conducted in the project. version 2: fix some typos, rephrase the definition, and add some more existing work

Multisensory Omni-directional Long-term Place Recognition: Benchmark Dataset and Analysis

Ashwin Mathur, Fei Han, Hao Zhang

专题命中 仿真评测 ：self-driving(abstract)；分类 cs.RO

Comments 15 pages

MultiCol-SLAM - A Modular Real-Time Multi-Camera SLAM System

Steffen Urban, Stefan Hinz

专题命中 仿真评测 ：self-driving(abstract)；分类 cs.CV

Comments 15 pages, 8 figures, 2 tables

Human Arm simulation for interactive constrained environment design

Liang Ma, Ruina Ma, Damien Chablat, Fouad Bennis

专题命中 仿真评测 ：trajectory planning(abstract)；分类 cs.RO

Comments International Journal on Interactive Design and Manufacturing (IJIDeM) (2012) 1-12. arXiv admin note: substantial text overlap with arXiv:1012.4327

PIPE: Personalizing Recommendations via Partial Evaluation

Naren Ramakrishnan

专题命中 仿真评测 ：BEV(abstract)；分类 cs.AI

Revolutionizing Validation and Verification: Explainable Testing Methodologies for Intelligent Automotive Decision-Making Systems

Halit Eris, Stefan Wagner

专题命中 仿真评测 ：autonomous driving(abstract,journal_ref)

Comments Preprint to be published at SE4ADS

Journal ref 2025 IEEE/ACM 1st International Workshop on Software Engineering for Autonomous Driving Systems (SE4ADS), Ottawa, ON, Canada, 2025, pp. 34-37

DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition

Marvin Chancán, Michael Milford

专题命中 仿真评测 ：分类 cs.RO、cs.CV、cs.AI；autonomous driving(journal_ref)

Comments 9 pages, 6 figures, 2 tables

Journal ref NeurIPS 2020 Workshop on Machine Learning for Autonomous Driving (ML4AD)

SCALE-Sim EVA: Design Principles for an Extensible, Visualizable, and Adaptable Accelerator Simulation Framework

SCALE-Sim EVA：可扩展、可可视化且可适配的加速器模拟框架的设计原则

Jingtian Dang, Ritik Raj, Tushar Krishna

专题命中 仿真评测 ：occupancy(abstract)

AI总结 SCALE-Sim EVA是一款面向IR感知加速器建模的可扩展、可可视化、可适配的模拟框架，通过张量命令与命令分解机制计算周期时序，可生成轨迹用于分析执行相关指标，解决了固定加速器模拟器难以扩展的问题。

From Annual Throughput to Vessel Schedules: A Stochastic Generator for Transshipment Hub Simulation

从年度吞吐量到船舶班期：一种用于中转枢纽仿真的随机生成器

Qiaohong Li, Haobin Li, Tianhao Chen, Ek Peng Chew

专题命中 仿真评测 ：occupancy(abstract)

AI总结 该研究针对集装箱中转枢纽仿真的现有模型预测失真问题，提出基于公共港口数据校准的Gamma-GDF随机生成器，经两大超级枢纽验证精度良好，为港口仿真提供稳健基础。

FAIR-Compute: A Roadmap for Fair and Efficient Allocation of Federated Digital Research Infrastructure

FAIR-Compute：联邦数字研究基础设施的公平与高效分配路线图

Konstantinos E. Zachariadis, Ahmed Sayed, Dimitris Fotakis, Angeliki Mathioudaki, Wan Shuen Siaw

专题命中 仿真评测 ：occupancy(abstract)

AI总结 FAIR-Compute 针对联邦数字研究基础设施的分配问题，结合多领域方法与实证研究，发现资源分配存在占用与利用率偏差、调度可优化、联邦需管控等关键问题，提出相关路线图方向。

Comments Final Report to the National Federated Compute Services (NFCS) Flexible Fund

Evaluating Fuzz Testing for Reinforcement Learning Agents

评估强化学习智能体的模糊测试

Zhibin Kang, Hanmo You, Dong Wang, Haiming Zheng, Junjie Chen

机构 * College of Intelligence and Computing, Tianjin University（天津大学智能与计算学部） 纠错

专题命中 仿真评测 ：autonomous driving(abstract)

AI总结 研究针对强化学习智能体模糊测试评估差异问题，通过在三种环境下统一配置对五种方法及随机测试进行基准测试，从多角度评估，揭示不同方法特点，表明模糊测试能提升智能体鲁棒性等，还给出可操作指导。

Transfer Learning Architectures for Scalable Multi-Fidelity Bayesian Optimization

用于可扩展多保真贝叶斯优化的迁移学习架构

Jaewook Lee, Ethan Errington, Christian D. Lorenz, Miao Guo

机构 * King’s College London（伦敦国王学院） 纠错

专题命中 仿真评测 ：self-driving(abstract)

AI总结 研究针对可扩展多保真贝叶斯优化，在相同条件下对11种迁移学习代理和4种GP方法进行基准测试，发现迁移学习代理在分子和材料问题上表现更佳，贪婪利用迁移学习均值是更稳健选择，是分子和材料MFBO的首选代理。

MC-BRIDGE: A Modular Receiver-Chain Simulation Framework for OECT-Based Molecular Communication

MC-BRIDGE：用于基于有机电化学晶体管的分子通信的模块化接收链仿真框架

Hongbin Ni, Ozgur B. Akan

专题命中 仿真评测 ：occupancy(abstract)

AI总结 研究基于OECT的分子通信，提出MC-BRIDGE模块化框架，连接多种过程到电荷域检测，能生成噪声、支持多种键控，通过公共接口估计相关指标并分析干扰，经测试得出满足SER目标的预算上限，揭示多种因素对接收器结论的影响。

Comments 13 pages, 7 figures, submitted to IEEE Transactions on Molecular, Biological, and Multi-Scale Communications (TMBMC)

HybridQC: Hardware-Grounded Simulation of Tightly Integrated Hybrid Quantum-Classical Systems

HybridQC：紧密集成的混合量子-经典系统的硬件基础模拟

Panayiotis Christou, Shuwen Kan, Ying Mao

专题命中 仿真评测 ：occupancy(abstract)

AI总结 研究混合量子-经典系统性能受经典控制等因素限制的问题，提出HybridQC模拟器，通过将HCU建模为可配置图、分解作业并校准测量，可评估混合架构多方面限制，为其性能评估提供系统框架。

The Precursor Genome: A Pairwise Reaction Dataset for Solid-State Synthesis

前驱体基因组：用于固态合成的成对反应数据集

Lauren N. Walters, Matthew J. McDermott, Bernardus Rendy, Yuxing Fei, Kristin A. Persson, Gerbrand Ceder

专题命中 仿真评测 ：self-driving(abstract)

AI总结 针对固态反应缺乏相关数据集阻碍合成科学发展的问题，通过A-Lab自动驾驶实验室生成前驱体基因组数据集，涵盖多种反应、元素及元数据，经自动精修和人工验证，为固态反应性预测模型提供可重复使用的基准。

Comments 4 figures, 20 pages, article submission

E-CoDrive: A Co-Simulation Framework for Testing Energy-Critical Driving Scenarios

E-CoDrive：用于测试能源关键型驾驶场景的联合仿真框架

Manfredi Napolitano, Alessandra Somma, Alessio Gambi, Andrea Stocco, Nicola Mazzocca

专题命中 仿真评测 ：autonomous driving(abstract)

AI总结 研究自动驾驶中能源消耗问题，核心方法是用E-CoDrive框架整合多种模拟器，贡献是可对自动驾驶新能源车软件栈进行基于能源感知场景测试，揭示交通条件对能耗的影响。

Comments In proceedings of the 41st IEEE/ACM International Conference on Automated Software Engineering (ASE '26) - Tools and Datasets track

The Calibration Turn in AI-Assisted Research: A Conceptual and Methodological Framework for Evidence-Licensed Claims

AI辅助研究中的校准转向：证据许可主张的概念与方法论框架

Hongmin Li

机构 * School of Life Science and Technology, Institute of Science Tokyo（东京科学大学生命科学与技术学院） ； Department of Computational Biology and Medical Sciences, Graduate School of Frontier Sciences, The University of Tokyo（东京大学前沿科学研究生院计算生物学与医学科学系） 纠错

专题命中 仿真评测 ：self-driving(abstract)

AI总结 提出AI辅助研究中证据许可主张的概念与方法论框架，通过五个算子描述研究过程，强调校准作为管理科学断言权的机制，并区分不同语义类型。

Comments 42 pages, 4 figures. Companion code and synthetic simulation artifacts: https://github.com/Li-Hongmin/calibration-turn-ai-assisted-research

Multiport Analytical Pixel Electromagnetic Simulator (MAPES) for AI-assisted RFIC and Microwave Circuit Design

多端口分析像素电磁模拟器（MAPES）用于AI辅助射频集成电路和微波电路设计

Junhui Rao, Yi Liu, Jichen Zhang, Zhaoyang Ming, Tianrui Qiao, Yujie Zhang, Chi Yuk Chiu, Hua Wang, Ross Murch

专题命中 仿真评测 ：occupancy(abstract)

AI总结 本文提出了一种新型分析框架MAPES，用于高效准确预测任意像素基微波和射频集成电路结构的电磁性能，通过虚拟像素和多层/通孔端口公式实现快速分析，无需额外全波计算，提升预测精度和效率。

Revisiting Action Factorization for Complex Action Spaces

重新审视复杂动作空间的动作分解

Timothy Flavin, Sandip Sen

机构 * The University of Tulsa（塔尔萨大学） 纠错

专题命中 仿真评测 ：autonomous driving(abstract)

AI总结 本文对多种动作分解方法在PPO、SAC、DQN算法和离散、混合、连续动作空间上进行了横截面研究，提出了VDN-PPO和PPO-MIX，并发现分支决斗架构在计算和性能之间取得了最佳平衡。

Comments 53 Pages, 37 Figures, 6 Tables, Target Journal/Venue: ACM Transactions on Autonomous and Adaptive Systems TAAS

SafeGen: LLM-Driven Assertion Generation and Fault Criticality Evaluation for Functional Safety

SafeGen: 面向功能安全的LLM驱动断言生成与故障关键性评估

Xuanyi Tan, Arjun Chaudhuri, Rubin Parekhji, Krishnendu Chakrabarty

专题命中 仿真评测 ：autonomous driving(abstract)

AI总结 提出SafeGen框架，利用大语言模型和超知识图谱从设计文档提取规范并生成功能安全断言，结合门级到RTL故障映射与形式验证实现故障关键性语义分级，在FOC系统中验证了优于传统方法。

Comments Accepted by DAC 2026

Constrained Co-evolutionary Metamorphic Differential Testing for Autonomous Systems with an Interpretability Approach

具有可解释性方法的约束共进化元形差分测试用于自主系统

Hossein Yousefizadeh, Shenghui Gu, Lionel C. Briand, Ali Nasr

专题命中 仿真评测 ：autonomous driving(abstract)

AI总结 本文提出CoCoMagic方法，结合元形测试、差分测试和先进搜索技术，通过约束共进化搜索生成可解释的测试用例，有效识别自主系统版本间的行为差异。

Comments Submitted to ACM Transactions on Software Engineering and Methodology journal(TOSEM)

Journal ref ACM Transactions on Software Engineering and Methodology, 2026

UniFFBench: Evaluating Universal Machine Learning Force Fields Against Experimental Measurements

评估通用机器学习力场与实验测量的对比

Sajid Mannan, Vaibhav Bihani, Carmelo Gonzales, Kin Long Kelvin Lee, Nitya Nand Gosvami, Sayan Ranu, Santiago Miret, N M Anoop Krishnan

机构 * Department of Civil Engineering, Indian Institute of Technology Delhi（印度理工学院德里土木工程系） ； Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi（印度理工学院德里人工智能学院） ； Intel Labs, California, USA（美国加州英特尔实验室） ； Department of Materials Science and Engineering, Indian Institute of Technology Delhi（印度理工学院德里材料科学与工程系） ； Department of Computer Science and Engineering, Indian Institute of Technology Delhi（印度理工学院德里计算机科学与工程系） 纠错

专题命中 仿真评测 ：occupancy(abstract)

AI总结 提出UniFFBench框架和MinX数据集，系统评估六种通用机器学习力场，发现模型在计算基准上表现优异但在实验复杂性下存在显著“现实差距”，密度预测误差高于实际应用阈值。