Yong Dai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yong Dai has a total of 90 co-authors. Here are the top co-authors by number of shared publications:
- Zenglin Xu
Shanghai Academy of AI for Science Shanghai China
4 shared publications - Che Liu
Imperial College London, London, United Kingdom
2 shared publications - Dong Zhang
HiThink Research, Hangzhou, China
2 shared publications - Haipang Wu
HiThink Research, Hangzhou, China
2 shared publications - Ji Liu
Baidu Inc., Beijing, China
2 shared publications Jiang Duan
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 14498-14506
Bridging VLMs and Embodied Intelligence with Deliberate Practice Policy Optimization
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 14498-14506
Open AccessBridging VLMs and Embodied Intelligence with Deliberate Practice Policy Optimization
- 18 February 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 339
MRDNet: Multivariable Relational Decomposition Network for Multivariate Time Series Forecasting
- 18 February 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 339
MRDNet: Multivariable Relational Decomposition Network for Multivariate Time Series Forecasting
- 5 February 2026
- Published by Elsevier in Pattern Recognition
- Vol. 176
VLCounting: Taming zero-shot counting via language-driven exemplar grounding
- 5 February 2026
- Published by Elsevier in Pattern Recognition
- Vol. 176
VLCounting: Taming zero-shot counting via language-driven exemplar grounding
- 20 November 2025
- Published by Elsevier in Neural Networks
- Vol. 196, 108312
TimeCNN: Refining inscross-variable interaction on time point for time series forecasting
- 20 November 2025
- Published by Elsevier in Neural Networks
- Vol. 196, 108312
TimeCNN: Refining inscross-variable interaction on time point for time series forecasting
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10787-10796
NEXUS-O: An Omni-Perceptive and -Interactive Model for Language, Audio, and Vision
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10787-10796
Open AccessNEXUS-O: An Omni-Perceptive and -Interactive Model for Language, Audio, and Vision
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12867-12874
MME-Finance: A Multimodal Finance Benchmark for Expert-level Understanding and Reasoning
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12867-12874
Open AccessMME-Finance: A Multimodal Finance Benchmark for Expert-level Understanding and Reasoning
- Open Access
- 29 April 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
TencentLLMEval: A Hierarchical Evaluation of Real-World Capabilities for Human-Aligned LLMs
Large language models (LLMs) have shown impressive capabilities across various natural language tasks. However, evaluating their alignment with human preferences remains a challenge. To this end, we propose a comprehensive human evaluation framework to assess LLMs’ proficiency in following instructions on diverse real-world tasks. We construct a hierarchical task tree encompassing 7 major areas covering over 200 categories and over 800 tasks, which covers diverse capabilities such as question answering, reasoning, multiturn dialogue, and text generation, to evaluate LLMs in a comprehensive and in-depth manner. We also design detailed evaluation standards and processes to facilitate consistent, unbiased judgments from human evaluators. A test set of over 3,000 instances is released, spanning different difficulty levels and knowledge domains. Our work provides a standardized methodology to evaluate human alignment in LLMs for both English and Chinese. We also analyze the feasibility of automating parts of evaluation with a strong LLM (GPT-4). Our framework supports a thorough assessment of LLMs as they are integrated into real-world applications. We have made publicly available the task tree, TencentLLMEval dataset, and evaluation methodology which have been demonstrated as effective in assessing the performance of Tencent Hunyuan LLMs 1 . By doing so, we aim to facilitate the benchmarking of advances in the development of safe and human-aligned LLMs.
- 29 April 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
Open AccessTencentLLMEval: A Hierarchical Evaluation of Real-World Capabilities for Human-Aligned LLMs
Large language models (LLMs) have shown impressive capabilities across various natural language tasks. However, evaluating their alignment with human preferences remains a challenge. To this end, we propose a comprehensive human evaluation framework to assess LLMs’ proficiency in following instructions on diverse real-world tasks. We construct a hierarchical task tree encompassing 7 major areas covering over 200 categories and over 800 tasks, which covers diverse capabilities such as question answering, reasoning, multiturn dialogue, and text generation, to evaluate LLMs in a comprehensive and in-depth manner. We also design detailed evaluation standards and processes to facilitate consistent, unbiased judgments from human evaluators. A test set of over 3,000 instances is released, spanning different difficulty levels and knowledge domains. Our work provides a standardized methodology to evaluate human alignment in LLMs for both English and Chinese. We also analyze the feasibility of automating parts of evaluation with a strong LLM (GPT-4). Our framework supports a thorough assessment of LLMs as they are integrated into real-world applications. We have made publicly available the task tree, TencentLLMEval dataset, and evaluation methodology which have been demonstrated as effective in assessing the performance of Tencent Hunyuan LLMs 1 . By doing so, we aim to facilitate the benchmarking of advances in the development of safe and human-aligned LLMs.
- 26 March 2025
- Published by Elsevier in Neurocomputing
- Vol. 637
ConceptPsy: A comprehensive benchmark suite for hierarchical psychological concept understanding in LLMs
- 26 March 2025
- Published by Elsevier in Neurocomputing
- Vol. 637
ConceptPsy: A comprehensive benchmark suite for hierarchical psychological concept understanding in LLMs
- 7 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1685-1692
Wind-storage Control Strategy Considering the Electro-Thermal Coupling Characteristics of Overhead Lines
When large-scale wind power is connected to the grid through overhead transmission lines (OTLs), it will cause the problem of power fluctuation and line transmission capacity exceeding the limit, which seriously endangers the stable operation of power system. In this paper, reasonable wind abandonment is carried out to meet the transmission capacity constraint based on electro-thermal coupling theory. Meanwhile, the output power after wind abandonment is stabilized by hybrid energy storage to further meet the requirements of wind power grid-connected volatility. First, the wind-storage integration model is established. Then, based on the electro-thermal coupling theory, a wind abandonment strategy is established to meet the transmission capacity constraints of the OTLs, which uses real-time meteorological parameters to calculate the standard of wind abandonment. Then, taking the output power after wind abandonment as input, a wind-storage control strategy based on hybrid energy storage is established, which can adjust wind power output to meet the constraints of wind power grid-connected volatility. Finally, case analysis shows that the proposed strategy can not only realize the safe grid-connection of wind power, but also reasonably distribute the hybrid energy storage power instructions.
- 7 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1685-1692
Wind-storage Control Strategy Considering the Electro-Thermal Coupling Characteristics of Overhead Lines
When large-scale wind power is connected to the grid through overhead transmission lines (OTLs), it will cause the problem of power fluctuation and line transmission capacity exceeding the limit, which seriously endangers the stable operation of power system. In this paper, reasonable wind abandonment is carried out to meet the transmission capacity constraint based on electro-thermal coupling theory. Meanwhile, the output power after wind abandonment is stabilized by hybrid energy storage to further meet the requirements of wind power grid-connected volatility. First, the wind-storage integration model is established. Then, based on the electro-thermal coupling theory, a wind abandonment strategy is established to meet the transmission capacity constraints of the OTLs, which uses real-time meteorological parameters to calculate the standard of wind abandonment. Then, taking the output power after wind abandonment as input, a wind-storage control strategy based on hybrid energy storage is established, which can adjust wind power output to meet the constraints of wind power grid-connected volatility. Finally, case analysis shows that the proposed strategy can not only realize the safe grid-connection of wind power, but also reasonably distribute the hybrid energy storage power instructions.
- 1 January 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 236
Graph Fusion Network for Text Classification
- 1 January 2022
- Published by Elsevier in Knowledge-Based Systems
- Vol. 236
Graph Fusion Network for Text Classification