让我们面对现实，让我们终于理想(2020/01/26)
Bingning Wang (王炳宁)
Bingning is currently the head of Pre-training at Baichuan Intelligence. He received his Ph.D from the Institute of Automation, Chinese Academy of Sciences in 2018, under the supervision of Prof. Jun Zhao and Kang Liu, with research focused on question answering systems and generative models.
Before joining Baichuan, he held senior research positions at Sogou and Tencent, accumulating extensive experience in large-scale generative models. He led the creation and release of several large-scale Chinese QA datasets, including ReCO, ComQA, ChiQA, and T2Ranking. At Tencent, he contributed to the development of the "Shenzhou" series of BERT-based language models, which achieved top rankings on the CLUE leaderboard.
He is the leading force behind the Baichuan series of pre-trained models, which have garnered over 10,000 stars on GitHub and more than 10 million downloads globally on Hugging Face. Wang has published 12 first-author papers in top-tier AI and NLP conferences such as ACL, SIGIR, and AAAI, and received the Best Paper runner-up Award at CIKM 2021. His Ph.D. dissertation, Key Technologies for Machine Reading Comprehension, was recognized as an Excellent Doctoral Dissertation by the Chinese Information Processing Society in 2019. He also serves as an executive member of the Youth Working Committee of the Chinese Information Processing Society.
Research Interests and Thoughts
What's New
- Feb, 2025: Two papers were accepted by CVPR 2025.
- Jan, 2025: Two papers were accepted by ICLR 2025.
- Jan, 2025: One paper was accepted by NAACL 2025.
- Oct, 2024: Two papers were accepted by EMNLP 2024.
- Oct, 2024: Two papers were accepted by NeurIPS 2024.
Projects
- Baichuan-m1: Pushing the medical capability of large language models. A 14B multi-lingual large language model dedicated for medical domain. Developed by my team at Baichuan Inc. GitHub, 🤗 Huggingface, Paper
- Baichuan-2: Open large-scale language models. A 7B and 13B bilingual large language model developed by Baichuan Inc. GitHub, 🤗 Huggingface, Paper
- T2Ranking: a large-scale Chinese benchmark for passage ranking. GitHub, 🤗 Huggingface, Paper
- ChiQA: A Large Scale Image-based Real-World Question Answering Dataset for Multi-Modal Understanding. GitHub, Paper
- An encoder-decoder based Chinese-English neural machine translation system based on vanilla pytorch (Its brittle, don't try to crush it). Demo, GitHub, Blog
Publications [Google Scholar]
- Wang Bingning; Zhao, Haizhou; Zhou, Huozhi; Song, Liang; Xu, Mingyu; Cheng, Wei; Zeng, Xiangrong; Zhang, Yupeng; Huo, Yuqi; Wang, Zecheng; others. “Baichuan-m1: Pushing the medical capability of large language models.” arxiv, 2025.
- Du, Yifan; Liu, Zikang; Li, Yifan; Zhao, Wayne Xin; Huo, Yuqi; Wang Bingning; Chen, Weipeng; Liu, Zheng; Wang, Zhongyuan; Wen, Ji-Rong. “Virgo: A Preliminary Exploration on Reproducing o1-like MLLM.” arxiv, 2025.
- Dong, Zican; Li, Junyi; Jiang, Jinhao; Xu, Mingyu; Zhao, Wayne Xin; Wang Bingning; Chen, Weipeng. “LongReD: Mitigating Short-Text Degradation of Long-Context Large Language Models via Restoration Distillation.” arxiv, 2025.
- Zhao, Zijia; Huo, Yuqi; Yue, Tongtian; Guo, Longteng; Lu, Haoyu; Wang Bingning; Chen, Weipeng; Liu, Jing. “Efficient Motion-Aware Video MLLM.” CVPR, 2025.
- Men, Xin; Xu, Mingyu; Zhang, Qingyu; Wang Bingning; Lin, Hongyu; Lu, Yaojie; Han, Xianpei; Chen, Weipeng. “Shortgpt: Layers in large language models are more redundant than you expect.” arxiv, 2024.
- Dong, Zican; Li, Junyi; Men, Xin; Zhao, Xin; Wang Bingning; Tian, Zhen; Wen, Ji-Rong; others. “Exploring context window of large language models via decomposed positional vectors.” Advances in Neural Information Processing Systems, 2024.
- Liu, Deyuan; Wang, Zecheng; Wang Bingning; Chen, Weipeng; Li, Chunshan; Tu, Zhiying; Chu, Dianhui; Li, Bo; Sui, Dianbo. “Checkpoint merging via bayesian optimization in llm pretraining.” arxiv, 2024.
- Zhao, Zijia; Lu, Haoyu; Huo, Yuqi; Du, Yifan; Yue, Tongtian; Guo, Longteng; Wang Bingning; Chen, Weipeng; Liu, Jing. “Needle in a video haystack: A scalable synthetic framework for benchmarking video mllms.” CVPR, 2025.
- Du, Yifan; Zhou, Kun; Huo, Yuqi; Li, Yifan; Zhao, Wayne Xin; Lu, Haoyu; Zhao, Zijia; Wang Bingning; Chen, Weipeng; Wen, Ji-Rong. “Towards event-oriented long video understanding.” ICLR, 2025.
- Liu, Han; Zhang, Yupeng; Wang Bingning; Chen, Weipeng; Hu, Xiaolin. “Full-ECE: A Metric For Token-level Calibration on Large Language Models.” arxiv, 2024.
- Chen, Jie; Zhang, Yupeng; Wang Bingning; Zhao, Wayne Xin; Wen, Ji-Rong; Chen, Weipeng. “Unveiling the flaws: exploring imperfections in synthetic data and mitigation strategies for large language models.” EMNLP, 2024.
- Liu, Han; Cui, Peng; Wang Bingning; Zhu, Jun; Hu, Xiaolin. “Accurate and Reliable Predictions with Mutual-Transport Ensemble.” arxiv, 2024.
- Zhou, Yuyan; Song, Liang; Wang Bingning; Chen, Weipeng. “Metagpt: Merging large language models using model exclusive task arithmetic.” EMNLP, 2024.
- Dong, Guosheng; Pan, Da; Sun, Yiding; Zhang, Shusen; Liang, Zheng; Wu, Xin; Shen, Yanjun; Yang, Fan; Sun, Haoze; Li, Tianpeng; others. “Baichuanseed: Sharing the potential of extensive data collection and deduplication by introducing a competitive large language model baseline.” arxiv, 2024.
- Du, Yifan; Huo, Yuqi; Zhou, Kun; Zhao, Zijia; Lu, Haoyu; Huang, Han; Zhao, Wayne Xin; Wang Bingning; Chen, Weipeng; Wen, Ji-Rong. “Exploring the design space of visual context representation in video mllms.” ICLR, 2025.
- Chen, Zhipeng; Song, Liang; Zhou, Kun; Zhao, Wayne Xin; Wang Bingning; Chen, Weipeng; Wen, Ji-Rong. “Extracting and Transferring Abilities For Building Multi-lingual Ability-enhanced Large Language Models.” arxiv, 2024.
- Huang, Han; Huo, Yuqi; Zhao, Zijia; Lu, Haoyu; Wu, Shu; Wang Bingning; Liu, Qiang; Chen, Weipeng; Wang, Liang. “Beyond Filtering: Adaptive Image-Text Quality Enhancement for MLLM Pretraining.” arxiv, 2024.
- Xu, Mingyu; Men, Xin; Wang Bingning; Zhang, Qingyu; Lin, Hongyu; Han, Xianpei; others. “Base of rope bounds context length.” The Thirty-eighth Annual Conference on Neural Information Processing Systems, 2024.
- Xu, Mingyu; Cheng, Wei; Wang Bingning; Chen, Weipeng. “KV Shifting Attention Enhances Language Modeling.” arxiv, 2024.
- Xu, Yao; He, Shizhu; Chen, Jiabei; Xiangrong, Zeng; Wang Bingning; Liu, Guang; Zhao, Jun; Liu, Kang. “Llasa: Large language and structured data assistant.” NAACL, 2025.
- Xie, Xiaohui; Dong, Qian; Wang Bingning; Lv, Feiyang; Yao, Ting; Gan, Weinan; Wu, Zhijing; Li, Xiangsheng; Li, Haitao; Liu, Yiqun; others. “T2ranking: A large-scale chinese benchmark for passage ranking.” Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2023.
- Yang, Aiyuan; Xiao, Bin; Wang Bingning; Zhang, Borong; Bian, Ce; Yin, Chao; Lv, Chenxu; Pan, Da; Wang, Dian; Yan, Dong; others. “Baichuan 2: Open large-scale language models.” arxiv, 2023. Corresponding Author.
- Liu, Han; Wang Bingning; Yao, Ting; Liang, Haijin; Xu, Jianjin; Hu, Xiaolin. “Bridging the Gap Between Training and Inference of Bayesian Controllable Language Models.” arxiv, 2022.
- Wang Bingning; Lv, Feiyang; Yao, Ting; Ma, Jin; Luo, Yu; Liang, Haijin. “Chiqa: A large scale image-based real-world question answering dataset for multi-modal understanding.” Proceedings of the 31st ACM International Conference on Information & Knowledge Management, 2022.
- Wang Bingning; Yao, Ting; Chen, Weipeng; Xu, Jingfang; Wang, Xiaochuan. “ComQA: Compositional question answering via hierarchical graph neural networks.” Proceedings of the Web Conference 2021, 2021.
- Wang Bingning; Yao, Ting; Chen, Weipeng; Xu, Jingfang; Wang, Xiaochuan. “Multi-lingual question generation with language agnostic language model.” Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, 2021.
- Yang, Han; Zhang, Leilei; Wang Bingning; Yao, Ting; Liu, Junfei. “Cycle or Minkowski: Which is more appropriate for knowledge graph embedding?” Proceedings of the 30th ACM International Conference on Information & Knowledge Management, 2021.
- Wang Bingning; Yao, Ting; Zhang, Qi; Xu, Jingfang; Wang, Xiaochuan. “Reco: A large scale chinese reading comprehension dataset on opinion.” Proceedings of the AAAI Conference on Artificial Intelligence, 2020.
- Wang Bingning; Wang, Xiaochuan; Tao, Ting; Zhang, Qi; Xu, Jingfang. “Neural question generation with answer pivot.” Proceedings of the AAAI conference on artificial intelligence, 2020.
- Xing, Xiaoyu; Jin, Zhijing; Jin, Di; Wang Bingning; Zhang, Qi; Huang, Xuanjing. “Tasty burgers, soggy fries: Probing aspect robustness in aspect-based sentiment analysis.” arxiv, 2020.
- Luo, Yichao; Li, Zhengyan; Wang Bingning; Xing, Xiaoyu; Zhang, Qi; Huang, Xuanjing. “SenSeNet: Neural keyphrase generation with document structure.” arxiv, 2020.
- Men, Xin; Kong, Xiangli; Yang, Xiaopan; Wang Bingning; Wang, Yue; Liu, Ya; Yu, Longbiao; Li, Hongliang; Xu, Binghui. “Synthesis of a pomegranate shaped reduced graphene oxide stabilized secondary Si nanoparticles composite anode for lithium ion batteries.” International Journal of Hydrogen Energy, 2020.
- Wu, Jindou; Yang, Yunlun; Deng, Chao; Tang, Hongyi; Wang Bingning; Sun, Haoze; Yao, Ting; Zhang, Qi. “Sogou machine reading comprehension toolkit.” arxiv, 2019.
- Wang Bingning; Yao, Ting; Zhang, Qi; Xu, Jingfang; Tian, Zhixing; Liu, Kang; Zhao, Jun. “Document gated reader for open-domain question answering.” Proceedings of the 42nd international ACM SIGIR conference on research and development in information retrieval, 2019.
- Wang Bingning; Yao, Ting; Zhang, Qi; Xu, Jingfang; Liu, Kang; Tian, Zhixing; Zhao, Jun. “Unsupervised story comprehension with hierarchical encoder-decoder.” Proceedings of the 2019 ACM SIGIR International Conference on Theory of Information Retrieval, 2019.
- Wang Bingning; Liu, Kang; Zhao, Jun. “Deep semantic hashing with multi-adversarial training.” Proceedings of the 27th ACM International Conference on Information and Knowledge Management, 2018.
- Wang Bingning; Liu, Kang; Zhao, Jun. “Conditional Generative Adversarial Networks for Commonsense Machine Comprehension.” IJCAI, 2017.
- Wang Bingning; Guo, Shangmin; Liu, Kang; He, Shizhu; Zhao, Jun. “Employing External Rich Knowledge for Machine Comprehension.” IJCAI, 2016.
- Wang Bingning; Liu, Kang; Zhao, Jun. “Inner attention based recurrent neural networks for answer selection.” Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2016.
Awards
- Best Paper Award at CIKM 2021.
- 2019 Excellent Doctoral Dissertation Award from the Chinese Society for Chinese Information Processing – “机器阅读理解关键技术研究”
- First Prize in the North China Division of the 2012 Freescale Intelligent Vehicle Competition.
Service
- Executive Committee Member of the Youth Working Committee of the Chinese Information Processing Society.
- Area Chair for ACL and EMNLP (Bigger Task)
Contact
Email: god@bingning.wang, daniel@baichuan-inc.com
Miscellaneous
- My MBTI is INFP.
- While I was dating my wife, there was a time when I was deeply into writing poetry. Here are a few pieces I wrote during that period [link].
- Kobe Bryant is my favorite athlete. He gave me immense strength during the most difficult times in my life. May he rest peacefully in heaven.
- I have a passion for capturing the night sky. Someday, I hope to gather my photographs into an album that tells my story through the stars.