🧑‍🔬 Yehui Tang

🏛️ Affiliations

🏢SamsungShanghai · China2026 · 1 papercurrent🏢Huawei TechnologiesShenzhen · China2020–2025 · 31 papers🎓Peking UniversityBeijing · China2019–2023 · 19 papers🏢Huawei TechnologiesMarkham · Canada2024 · 2 papers🎓The University of SydneySydney · Australia2024 · 2 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 2 more institutions on single bylines are not listed.

🧭 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📅 Works by year

2020s

📄 Papers

The most cited first (OpenAlex and our library).

1. Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM Reasoning (2024 · 43 citations)
   Prompting schemes such as Chain-of-Thought and Tree-of-Thought usually follow one reasoning attempt to its end.
2. PanGu-π: Enhancing Language Model Architectures via Nonlinearity Compensation (2023 · 10 citations)
   Progress in large language models has mostly come from adding parameters and data. The design of the Transformer block itself gets little attention, even though large models are costly to run. The authors study LLaMA-style decoders and identify feature collapse: a rank metric shows token representations growing more…
3. Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUs (2025 · 8 citations)
   Training MoE language models with nearly a trillion parameters is hard: the hardware is underused because the set of experts each token activates changes at runtime, and the scale strains communication and memory.
4. PanGu-π Pro:Rethinking Optimization and Architecture for Tiny Language Models (2024 · 8 citations)
   On-device deployment needs small language models that still perform well. Most tiny models, however, simply reuse recipes designed for much larger LLMs, and few studies test which of those choices actually suit a ~1B-parameter model. The authors run a systematic set of ablations on a 1B LLaMA-style baseline, trained…
5. Pangu Ultra: Pushing the Limits of Dense Large Language Models on Ascend NPUs (2025 · 6 citations)
   Pangu Ultra is Huawei's effort to show that a very deep dense Transformer, trained entirely on Ascend NPUs instead of GPUs, can compete with much larger sparse MoE systems.
6. Pangu Pro MoE: Mixture of Grouped Experts for Efficient Sparsity (2025 · 4 citations)
   Sparse Mixture-of-Experts LLMs are cheap per token. However, when experts are spread across accelerators, uneven routing overloads some devices, and the slowest device limits overall speed. The Pangu Team at Huawei proposes Mixture of Grouped Experts (MoGE). Experts are split into equal, non-overlapping groups…
7. Multi-Granularity Semantic Revision for Large Language Model Distillation (2026 · 2 citations)
   The paper addresses white-box knowledge distillation (KD) for auto-regressive LLMs, where a compact student learns from the full output distributions of a larger teacher.

All works

11 works of theirs in our library, the most cited first; 7 with a page here, the others link their original until we summarise them.

1. CMT: Convolutional Neural Networks Meet Vision Transformers ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetJianyuan Guo et al.2022 · 969 citations
2. GhostNetV2: Enhance Cheap Operation with Long-Range Attention ↗neural information processing systems · no summary yetYehui Tang et al.2022 · 340 citations
3. Vision GNN: An Image is Worth Graph of Nodes ↗neural information processing systems · no summary yetKai Han et al.2022 · 256 citations
4. Patch Slimming for Efficient Vision Transformers ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetYehui Tang et al.2022 · 204 citations
5. Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM ReasoningarXiv:2412.09078Zhenni Bi, Kai Han, Chuanjian Liu, et al.2024 · 43 citations
6. PanGu-π: Enhancing Language Model Architectures via Nonlinearity CompensationarXiv:2312.17276Yunhe Wang, Hanting Chen, Yehui Tang, et al.2023 · 10 citations
7. Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUsarXiv:2505.04519Yehui Tang, Yichun Yin, Yaoyuan Wang, et al.2025 · 8 citations
8. PanGu-π Pro:Rethinking Optimization and Architecture for Tiny Language ModelsarXiv:2402.02791Yehui Tang, Kai Han, Fangcheng Liu, et al.2024 · 8 citations
9. Pangu Ultra: Pushing the Limits of Dense Large Language Models on Ascend NPUsarXiv:2504.07866Yichun Yin, Wenyong Huang, Kaikai Song, et al.2025 · 6 citations
10. Pangu Pro MoE: Mixture of Grouped Experts for Efficient SparsityarXiv:2505.21411Yehui Tang, Xiaosong Li, Fangcheng Liu, et al.2025 · 4 citations
11. Multi-Granularity Semantic Revision for Large Language Model DistillationAnnual Meeting of the Association for Computational Linguistics (ACL)Xiaoyu Liu, Yun Zhang, Wei Li, et al.2026 · 2 citations

🤝 Co-authors here

12 of 25 co-authors with a page, the most papers together first.

• Fangcheng Liu ×4
• Kai Han ×4
• Yunhe Wang ×4
• Wei Guo ×3
• Yaoyuan Wang ×3
• Hang Zhou ×2
• Hanting Chen ×2
• Xiaojun Meng ×2
• Yichun Yin ×2
• Zheyuan Bai ×2
• Can Chen
• Dong Li

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

• IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings ×2
• neural information processing systems ×2
• Annual Meeting of the Association for Computational Linguistics (ACL): ACL 2026

🔗 Cited by

Papers of our library cite their works 142 times (of the 149,439 papers whose bibliographies we read); 27 citing papers with a page here →.

Updated 31 October 2026

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/authors/yehui-tang/