大模型对齐与安全 - arXivDaily 专题

2210.00074 2022-10-17 cs.LG stat.AP 57%

Leveraging Industry 4.0 -- Deep Learning, Surrogate Model and Transfer Learning with Uncertainty Quantification Incorporated into Digital Twin for Nuclear System

M. Rahman, Abid Khan, Sayeed Anowar, Md Al-Imran, Richa Verma, Dinesh Kumar, Kazuma Kobayashi, Syed Alam

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

2207.02440 2022-10-17 cs.LG stat.ML 57%

PAC Prediction Sets for Meta-Learning

Sangdon Park, Edgar Dobriban, Insup Lee, Osbert Bastani

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments Accepted to NeurIPS 2022

2206.01323 2022-10-14 cs.LG eess.SP 57%

SPD domain-specific batch normalization to crack interpretable unsupervised domain adaptation in EEG

Reinmar J Kobler, Jun-ichiro Hirayama, Qibin Zhao, Motoaki Kawanabe

专题命中幻觉与事实性：alignment(abstract)；分类 cs.LG

Comments 10 pages, accepted at NeurIPS 2022

2208.07464 2022-09-28 cs.LG math.OC stat.ML 57%

An Overview and Prospective Outlook on Robust Training and Certification of Machine Learning Models

Brendon G. Anderson, Tanmay Gautam, Somayeh Sojoudi

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

2208.01705 2022-08-09 cs.LG 57%

Success of Uncertainty-Aware Deep Models Depends on Data Manifold Geometry

Mark Penrod, Harrison Termotto, Varshini Reddy, Jiayu Yao, Finale Doshi-Velez, Weiwei Pan

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Journal ref International Conference on Machine Learning. PMLR 162 (2022)

2207.13341 2022-07-28 cs.LG 57%

Towards Clear Expectations for Uncertainty Estimation

Victor Bouvier, Simona Maggio, Alexandre Abraham, Léo Dreyfus-Schmidt

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

2207.03324 2022-07-08 cs.LG 57%

Calibrate to Interpret

Gregory Scafarto, Nicolas Posocco, Antoine Bonnefoy

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

Comments 16 pages, 9 figures, accepted at ECML PKDD 2022

2206.10718 2022-06-23 physics.comp-ph cs.LG physics.geo-ph 57%

Physics-informed machine learning with differentiable programming for heterogeneous underground reservoir pressure management

Aleksandra Pachalieva, Daniel O'Malley, Dylan Robert Harp, Hari Viswanathan

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

Comments 12 pages, 5 figures

2201.08418 2022-06-03 eess.IV cs.AI cs.CV physics.med-ph 57%

SoftDropConnect (SDC) -- Effective and Efficient Quantification of the Network Uncertainty in Deep MR Image Analysis

Qing Lyu, Christopher T. Whitlow, Ge Wang

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.AI

2205.15265 2022-05-31 cs.LG cs.CV 57%

Going Beyond One-Hot Encoding in Classification: Can Human Uncertainty Improve Model Performance?

Christoph Koller, Göran Kauermann, Xiao Xiang Zhu

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

2204.00382 2022-05-12 cs.LG cs.CV 57%

Autoencoder Attractors for Uncertainty Estimation

Steve Dias Da Cruz, Bertram Taetz, Thomas Stifter, Didier Stricker

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments This paper is accepted at IEEE International Conference on Pattern Recognition (ICPR), 2022

2204.13963 2022-05-02 cs.LG 57%

Tailored Uncertainty Estimation for Deep Learning Systems

Joachim Sicking, Maram Akila, Jan David Schneider, Fabian Hüger, Peter Schlicht, Tim Wirtz, Stefan Wrobel

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

2203.06290 2022-03-15 cs.LG cs.SY eess.SY math.OC 57%

SOCKS: A Stochastic Optimal Control and Reachability Toolbox Using Kernel Methods

Adam J. Thorpe, Meeko M. K. Oishi

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

2104.06135 2022-02-25 cs.LG cs.NE stat.ML 57%

Multivariate Deep Evidential Regression

Nis Meinert, Alexander Lavin

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments 20 pages, 13 figures

2112.09196 2022-01-26 cs.LG 57%

Benchmarking Uncertainty Quantification on Biosignal Classification Tasks under Dataset Shift

Tong Xia, Jing Han, Cecilia Mascolo

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

Comments Accepted by The 6th International Workshop on Health Intelligence (W3PHIAI-22)

2201.05669 2022-01-19 q-bio.QM cs.LG 57%

Prediction of Drug-Induced TdP Risks Using Machine Learning and Rabbit Ventricular Wedge Assay

Nan Miles Xi, Dalong Patrick Huang

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments arXiv admin note: text overlap with arXiv:2108.00543

2102.12680 2021-12-28 cs.LG 57%

Confidence Calibration with Bounded Error Using Transformations

Sooyong Jang, Radoslav Ivanov, Insup Lee, James Weimer

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

2112.06986 2021-12-21 cs.LG 57%

On The Reliability Of Machine Learning Applications In Manufacturing Environments

Nicolas Jourdan, Sagar Sen, Erik Johannes Husom, Enrique Garcia-Ceja, Tobias Biegel, Joachim Metternich

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments Workshop on Distribution Shifts, 35th Conference on Neural Information Processing Systems (NeurIPS 2021)

2111.09808 2021-11-19 cs.LG cs.CV 57%

Exploring the Limits of Epistemic Uncertainty Quantification in Low-Shot Settings

Matias Valdenegro-Toro

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments 7 pages, 3 figures, with supplementary material. LatinX in AI Research Workshop @ NeurIPS 2021

1911.07391 2021-11-16 cs.LG stat.ML 57%

Justification-Based Reliability in Machine Learning

Nurali Virani, Naresh Iyer, Zhaoyuan Yang

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments Extended version of paper accepted at AAAI 2020 with supplementary materials, update remark and fix typo

2110.13947 2021-10-28 cs.CV cs.LG 57%

Collaborative Uncertainty in Multi-Agent Trajectory Forecasting

Bohan Tang, Yiqi Zhong, Ulrich Neumann, Gang Wang, Ya Zhang, Siheng Chen

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments This paper has been accepted by NeurIPS 2021

2110.03553 2021-10-08 cs.AR cs.LG 57%

Shift-BNN: Highly-Efficient Probabilistic Bayesian Neural Network Training via Memory-Friendly Pattern Retrieving

Qiyu Wan, Haojun Xia, Xingyao Zhang, Lening Wang, Shuaiwen Leon Song, Xin Fu

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments 54th IEEE/ACM International Symposium on Microarchitecture

2109.04432 2021-09-10 cs.LG cs.IR stat.ML 57%

Detecting and Mitigating Test-time Failure Risks via Model-agnostic Uncertainty Learning

Preethi Lahoti, Krishna P. Gummadi, Gerhard Weikum

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.LG

Comments To appear in the 21st IEEE International Conference on Data Mining (ICDM 2021), Auckland, New Zealand

2108.08712 2021-08-20 cs.LG stat.ML 57%

Teaching Uncertainty Quantification in Machine Learning through Use Cases

Matias Valdenegro-Toro

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments 2nd Teaching in Machine Learning Workshop, Camera Ready, 5 pages, 3 figures

2107.11750 2021-08-02 cs.LG cs.CV 57%

Improving Variational Autoencoder based Out-of-Distribution Detection for Embedded Real-time Applications

Yeli Feng, Daniel Jun Xian Ng, Arvind Easwaran

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments This article appears as part of the ESWEEK-TECS special issue and will be presented in the International Conference on Embedded Software (EMSOFT), 2021

2107.14261 2021-08-02 physics.acc-ph cs.LG 57%

Quantifying Uncertainty for Machine Learning Based Diagnostic

Owen Convery, Lewis Smith, Yarin Gal, Adi Hanuka

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments arXiv admin note: substantial text overlap with arXiv:2105.04654

1910.09457 2021-06-24 cs.LG stat.ML 57%

Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods

Eyke Hüllermeier, Willem Waegeman

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments 59 pages

2010.12007 2021-06-16 cs.LG 57%

PRANK: motion Prediction based on RANKing

Yuriy Biktairov, Maxim Stebelev, Irina Rudenko, Oleh Shliazhko, Boris Yangel

专题命中幻觉与事实性：safety(abstract)；分类 cs.LG

Comments Accepted to NeurIPS 2020

2106.01410 2021-06-07 cs.AI 57%

Uncertainty Quantification 360: A Holistic Toolkit for Quantifying and Communicating the Uncertainty of AI

Soumya Ghosh, Q. Vera Liao, Karthikeyan Natesan Ramamurthy, Jiri Navratil, Prasanna Sattigeri, Kush R. Varshney, Yunfeng Zhang

专题命中幻觉与事实性：trustworthy(abstract)；分类 cs.AI

Comments Added references

2106.01170 2021-06-03 cs.CL 57%

Detecting Bot-Generated Text by Characterizing Linguistic Accommodation in Human-Bot Interactions

Paras Bhatt, Anthony Rios

专题命中幻觉与事实性：alignment(abstract)；分类 cs.CL

Comments 13 pages, to be published in Findings of ACL-IJCNLP 2021

1. 幻觉与事实性 1756 篇

Leveraging Industry 4.0 -- Deep Learning, Surrogate Model and Transfer Learning with Uncertainty Quantification Incorporated into Digital Twin for Nuclear System

PAC Prediction Sets for Meta-Learning

SPD domain-specific batch normalization to crack interpretable unsupervised domain adaptation in EEG

An Overview and Prospective Outlook on Robust Training and Certification of Machine Learning Models

Success of Uncertainty-Aware Deep Models Depends on Data Manifold Geometry

Towards Clear Expectations for Uncertainty Estimation

Calibrate to Interpret

Physics-informed machine learning with differentiable programming for heterogeneous underground reservoir pressure management

SoftDropConnect (SDC) -- Effective and Efficient Quantification of the Network Uncertainty in Deep MR Image Analysis

Going Beyond One-Hot Encoding in Classification: Can Human Uncertainty Improve Model Performance?

Autoencoder Attractors for Uncertainty Estimation

Tailored Uncertainty Estimation for Deep Learning Systems

SOCKS: A Stochastic Optimal Control and Reachability Toolbox Using Kernel Methods

Multivariate Deep Evidential Regression

Benchmarking Uncertainty Quantification on Biosignal Classification Tasks under Dataset Shift

Prediction of Drug-Induced TdP Risks Using Machine Learning and Rabbit Ventricular Wedge Assay

Confidence Calibration with Bounded Error Using Transformations

On The Reliability Of Machine Learning Applications In Manufacturing Environments

Exploring the Limits of Epistemic Uncertainty Quantification in Low-Shot Settings

Justification-Based Reliability in Machine Learning

Collaborative Uncertainty in Multi-Agent Trajectory Forecasting

Shift-BNN: Highly-Efficient Probabilistic Bayesian Neural Network Training via Memory-Friendly Pattern Retrieving

Detecting and Mitigating Test-time Failure Risks via Model-agnostic Uncertainty Learning

Teaching Uncertainty Quantification in Machine Learning through Use Cases

Improving Variational Autoencoder based Out-of-Distribution Detection for Embedded Real-time Applications

Quantifying Uncertainty for Machine Learning Based Diagnostic

Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods

PRANK: motion Prediction based on RANKing

Uncertainty Quantification 360: A Holistic Toolkit for Quantifying and Communicating the Uncertainty of AI

Detecting Bot-Generated Text by Characterizing Linguistic Accommodation in Human-Bot Interactions