大模型对齐与安全 - arXivDaily 专题

2402.17393 2024-02-28 cs.CY 57%

Designing Chatbots to Support Victims and Survivors of Domestic Abuse

Rahime Belen Saglam, Jason R. C. Nurse, Lisa Sugiura

专题命中安全训练：safety(abstract)；分类 cs.CY

2402.16569 2024-02-28 cs.CV cs.LG 57%

Pretrained Visual Uncertainties

Michael Kirchhof, Mark Collier, Seong Joon Oh, Enkelejda Kasneci

专题命中安全训练：trustworthy(abstract)；分类 cs.LG

2402.16036 2024-02-27 cs.RO cs.CV cs.LG 57%

Machine Learning-Based Vehicle Intention Trajectory Recognition and Prediction for Autonomous Driving

Hanyi Yu, Shuning Huo, Mengran Zhu, Yulu Gong, Yafei Xiang

专题命中安全训练：safety(abstract)；分类 cs.LG

2402.13408 2024-02-22 cs.CL 57%

Healthcare Copilot: Eliciting the Power of General LLMs for Medical Consultation

Zhiyao Ren, Yibing Zhan, Baosheng Yu, Liang Ding, Dacheng Tao

专题命中安全训练：safety(abstract)；分类 cs.CL

2311.01602 2024-02-20 cs.RO cs.AI 57%

DRNet: A Decision-Making Method for Autonomous Lane Changingwith Deep Reinforcement Learning

Kunpeng Xu, Lifei Chen, Shengrui Wang

专题命中安全训练：safety(abstract)；分类 cs.AI

2402.09387 2024-02-15 physics.plasm-ph cs.LG 57%

Active Disruption Avoidance and Trajectory Design for Tokamak Ramp-downs with Neural Differential Equations and Reinforcement Learning

Allen M. Wang, Oswin So, Charles Dawson, Darren T. Garnier, Cristina Rea, Chuchu Fan

专题命中安全训练：safety(abstract)；分类 cs.LG

2402.08968 2024-02-15 cs.AI 57%

GrounDial: Human-norm Grounded Safe Dialog Response Generation

Siwon Kim, Shuyang Dai, Mohammad Kachuee, Shayan Ray, Tara Taghavi, Sungroh Yoon

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments Accepted to findings of EACL 2024

2401.10949 2024-01-26 cs.MA cs.LG cs.SY eess.SY 57%

The Synergy Between Optimal Transport Theory and Multi-Agent Reinforcement Learning

Ali Baheri, Mykel J. Kochenderfer

专题命中安全训练：alignment(abstract)；分类 cs.LG

2401.05876 2024-01-12 cs.LG cs.RO 57%

Safe reinforcement learning in uncertain contexts

Dominik Baumann, Thomas B. Schön

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments Accepted final version to appear in the IEEE Transactions on Robotics

2312.11314 2023-12-19 cs.LG cs.LO cs.SY eess.SY 57%

Safeguarded Progress in Reinforcement Learning: Safe Bayesian Exploration for Control Policy Synthesis

Rohan Mitta, Hosein Hasanbeig, Jun Wang, Daniel Kroening, Yiannis Kantaros, Alessandro Abate

专题命中安全训练：safety(abstract)；分类 cs.LG

2312.05890 2023-12-12 cs.AI 57%

Scaling #DNN-Verification Tools with Efficient Bound Propagation and Parallel Computing

Luca Marzari, Gabriele Roncolato, Alessandro Farinelli

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments Accepted at AIRO 2023 the 10th Italian Workshop on Artificial Intelligence and Robotics co-located with the 22nd International Conference of the Italian Association for Artificial Intelligence (AI*IA 2023), Rome, Italy

2312.03296 2023-12-07 cs.RO cs.CV cs.LG 57%

Cooperative Probabilistic Trajectory Forecasting under Occlusion

Anshul Nayak, Azim Eskandarian

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 10 pages, 13 figures, 1 table

2312.01350 2023-12-05 cs.AI 57%

Honesty Is the Best Policy: Defining and Mitigating AI Deception

Francis Rhys Ward, Francesco Belardinelli, Francesca Toni, Tom Everitt

专题命中安全训练：safety(abstract)；分类 cs.AI

Comments Accepted as a spotlight at the 37th Conference on Neural Information Processing Systems (NeurIPS 2023)

2311.10538 2023-12-05 cs.AI 57%

Testing Language Model Agents Safely in the Wild

Silen Naihin, David Atkinson, Marc Green, Merwane Hamadi, Craig Swift, Douglas Schonholtz, Adam Tauman Kalai, David Bau

专题命中安全训练：safety(abstract)；分类 cs.AI

2205.06750 2023-11-21 cs.LG 57%

Provably Safe Reinforcement Learning: Conceptual Analysis, Survey, and Benchmarking

Hanna Krasowski, Jakob Thumm, Marlon Müller, Lukas Schäfer, Xiao Wang, Matthias Althoff

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments The published paper is available at https://openreview.net/forum?id=mcN0ezbnzO

Journal ref Transactions on Machine Learning Research, 2023

2311.07426 2023-11-14 cs.LG cs.CV cs.HC 57%

Optimising Human-AI Collaboration by Learning Convincing Explanations

Alex J. Chan, Alihan Huyuk, Mihaela van der Schaar

专题命中安全训练：safety(abstract)；分类 cs.LG

2302.00171 2023-11-02 cs.RO cs.LG cs.SY eess.SY math.OC 57%

Active Uncertainty Reduction for Safe and Efficient Interaction Planning: A Shielding-Aware Dual Control Approach

Haimin Hu, David Isele, Sangjae Bae, Jaime F. Fisac

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments The International Journal of Robotics Research. arXiv admin note: text overlap with arXiv:2202.07720

2302.06541 2023-10-24 cs.CL 57%

Towards Agile Text Classifiers for Everyone

Maximilian Mozes, Jessica Hoffmann, Katrin Tomanek, Muhamed Kouate, Nithum Thain, Ann Yuan, Tolga Bolukbasi, Lucas Dixon

专题命中安全训练：safety(abstract)；分类 cs.CL

Comments Findings of EMNLP 2023

2310.12432 2023-10-20 cs.LG 57%

CAT: Closed-loop Adversarial Training for Safe End-to-End Driving

Linrui Zhang, Zhenghao Peng, Quanyi Li, Bolei Zhou

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments 7th Conference on Robot Learning (CoRL 2023)

2212.07850 2023-10-19 cs.CL 57%

Attention as a Guide for Simultaneous Speech Translation

Sara Papi, Matteo Negri, Marco Turchi

专题命中安全训练：alignment(abstract)；分类 cs.CL

Comments Accepted to ACL 2023

Journal ref Proceedings of ACL 2023

2301.13807 2023-10-17 cs.SE cs.LG cs.NE cs.RO 57%

Identifying the Hazard Boundary of ML-enabled Autonomous Systems Using Cooperative Co-Evolutionary Search

Sepehr Sharifi, Donghwan Shin, Lionel C. Briand, Nathan Aschbacher

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments Accepted for publication by IEEE Transactions on Software Engineering (TSE)

2310.09426 2023-10-17 cs.LG stat.ML 57%

Offline Reinforcement Learning for Optimizing Production Bidding Policies

Dmytro Korenkevych, Frank Cheng, Artsiom Balakir, Alex Nikulkov, Lingnan Gao, Zhihao Cen, Zuobing Xu, Zheqing Zhu

专题命中安全训练：safety(abstract)；分类 cs.LG

2310.09360 2023-10-17 cs.LG 57%

Exact Verification of ReLU Neural Control Barrier Functions

Hongchao Zhang, Junlin Wu, Yevgeniy Vorobeychik, Andrew Clark

专题命中安全训练：safety(abstract)；分类 cs.LG

2309.15252 2023-09-28 cs.RO cs.LG 57%

V2X-Lead: LiDAR-based End-to-End Autonomous Driving with Vehicle-to-Everything Communication Integration

Zhiyun Deng, Yanjun Shi, Weiming Shen

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments To be published in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023

2309.14003 2023-09-26 cs.LG cs.RO 57%

Hierarchical Imitation Learning for Stochastic Environments

Maximilian Igl, Punit Shah, Paul Mougin, Sirish Srinivasan, Tarun Gupta, Brandyn White, Kyriacos Shiarlis, Shimon Whiteson

专题命中安全训练：safety(abstract)；分类 cs.LG

Comments Published at IROS'23

2309.12560 2023-09-25 cs.RO cs.AI 57%

Machine Learning Meets Advanced Robotic Manipulation

Saeid Nahavandi, Roohallah Alizadehsani, Darius Nahavandi, Chee Peng Lim, Kevin Kelly, Fernando Bello

专题命中安全训练：safety(abstract)；分类 cs.AI

2307.05812 2023-09-13 eess.SY cs.LG cs.SY 57%

Safe Reinforcement Learning for Strategic Bidding of Virtual Power Plants in Day-Ahead Markets

Ognjen Stanojev, Lesia Mitridati, Riccardo de Nardis di Prata, Gabriela Hug

专题命中安全训练：safety(abstract)；分类 cs.LG

2309.02812 2023-09-07 cs.CL 57%

Agent-based simulation of pedestrians' earthquake evacuation; application to Beirut, Lebanon

Rouba Iskandar, Kamel Allaw, Julie Dugdale, Elise Beck, Jocelyne Adjizian-Gérard, Cécile Cornou, Jacques Harb, Pascal Lacroix, Nada Badaro-Saliba, Stéphane Cartier, Rita Zaarour

专题命中安全训练：safety(abstract)；分类 cs.CL

2309.02328 2023-09-06 cs.RO cs.AI cs.SY eess.SY stat.ML 57%

Neurosymbolic Meta-Reinforcement Lookahead Learning Achieves Safe Self-Driving in Non-Stationary Environments

Haozhe Lei, Quanyan Zhu

专题命中安全训练：safety(abstract)；分类 cs.AI

2309.01918 2023-09-06 cs.RO cs.LG 57%

RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunking

Homanga Bharadhwaj, Jay Vakil, Mohit Sharma, Abhinav Gupta, Shubham Tulsiani, Vikash Kumar

专题命中安全训练：safety(abstract)；分类 cs.LG

1. 安全训练 3281 篇

Designing Chatbots to Support Victims and Survivors of Domestic Abuse

Pretrained Visual Uncertainties

Machine Learning-Based Vehicle Intention Trajectory Recognition and Prediction for Autonomous Driving

Healthcare Copilot: Eliciting the Power of General LLMs for Medical Consultation

DRNet: A Decision-Making Method for Autonomous Lane Changingwith Deep Reinforcement Learning

Active Disruption Avoidance and Trajectory Design for Tokamak Ramp-downs with Neural Differential Equations and Reinforcement Learning

GrounDial: Human-norm Grounded Safe Dialog Response Generation

The Synergy Between Optimal Transport Theory and Multi-Agent Reinforcement Learning

Safe reinforcement learning in uncertain contexts

Safeguarded Progress in Reinforcement Learning: Safe Bayesian Exploration for Control Policy Synthesis

Scaling #DNN-Verification Tools with Efficient Bound Propagation and Parallel Computing

Cooperative Probabilistic Trajectory Forecasting under Occlusion

Honesty Is the Best Policy: Defining and Mitigating AI Deception

Testing Language Model Agents Safely in the Wild

Provably Safe Reinforcement Learning: Conceptual Analysis, Survey, and Benchmarking

Optimising Human-AI Collaboration by Learning Convincing Explanations

Active Uncertainty Reduction for Safe and Efficient Interaction Planning: A Shielding-Aware Dual Control Approach

Towards Agile Text Classifiers for Everyone

CAT: Closed-loop Adversarial Training for Safe End-to-End Driving

Attention as a Guide for Simultaneous Speech Translation

Identifying the Hazard Boundary of ML-enabled Autonomous Systems Using Cooperative Co-Evolutionary Search

Offline Reinforcement Learning for Optimizing Production Bidding Policies

Exact Verification of ReLU Neural Control Barrier Functions

V2X-Lead: LiDAR-based End-to-End Autonomous Driving with Vehicle-to-Everything Communication Integration

Hierarchical Imitation Learning for Stochastic Environments

Machine Learning Meets Advanced Robotic Manipulation

Safe Reinforcement Learning for Strategic Bidding of Virtual Power Plants in Day-Ahead Markets

Agent-based simulation of pedestrians' earthquake evacuation; application to Beirut, Lebanon

Neurosymbolic Meta-Reinforcement Lookahead Learning Achieves Safe Self-Driving in Non-Stationary Environments

RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunking