🧑‍🔬 Yichun Yin

🏛️ Affiliations

🎓North China Electric Power UniversityBeijing · China2024–2026 · 8 paperscurrent🔬State Key Laboratory of Alternate Electrical Power System with Renewable Energy SourcesBeijing2025–2026 · 7 paperscurrent🎓Monash UniversityMelbourne · Australia2016–2020 · 17 papers🏢Huawei TechnologiesShenzhen · China2021–2023 · 4 papers🔬ARC Centre of Excellence for Electromaterials ScienceWollongong · Australia2016–2018 · 3 papers🎓Dongguan University of TechnologyDongguan · China2018–2019 · 2 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 3 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. TinyBERT: Distilling BERT for Natural Language Understanding (2020 · 2,378 citations)
   Large pre-trained language models such as BERT achieve very high performance. However, they have many parameters and slow inference, which makes them difficult to run on phones and other low-resource devices. The authors compress BERT with a knowledge distillation method designed specifically for the Transformer…
2. TernaryBERT: Distillation-aware Ultra-low Bit BERT (2020 · 213 citations)
   Fine-tuned BERT models are large: BERT-base has 109M parameters, over 400 MB in 32-bit floats. This makes them hard to run on phones and other constrained hardware. Earlier quantization work kept accuracy at 8 bits. Going lower usually needed mixed precision, which some hardware handles poorly, or product…
3. DQ-LoRe: Dual Queries with Low Rank Approximation Re-ranking for In-Context Learning (2023 · 27 citations)
   The order and choice of demonstrations in a prompt strongly affect how well large language models solve multi-step reasoning problems.
4. 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.
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. MATCH: Modulating Attention via In-Context Retrieval for Long-Context Transformers (2026)
   Sparse-attention LLMs, such as models built on sliding windows, save compute and memory. They tend to fail, however, when an answer depends on recalling an exact detail far back in the input. MATCH tackles this by pairing a sparsified Transformer with a lightweight retrieval component that treats the prompt itself as…

All works

11 works of theirs in our library, the most cited first; 6 with a page here, the others link their original until we summarise them.

1. TinyBERT: Distilling BERT for Natural Language UnderstandingFindings of the Association for Computational LinguisticsXiaoqi Jiao, Yichun Yin, Lifeng Shang, et al.2020 · 2,378 citations
2. TernaryBERT: Distillation-aware Ultra-low Bit BERTConference on Empirical Methods in Natural Language Processing (EMNLP)Wei Zhang, Lu Hou, Yichun Yin, et al.2020 · 213 citations
3. Generate & Rank: A Multi-task Framework for Math Word Problems ↗Findings of the Association for Computational Linguistics: EMNLP 2021 · no summary yetJianhao Shen et al.2021 · 152 citations
4. Document-Level Multi-Aspect Sentiment Classification as Machine Comprehension ↗Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · no summary yetYichun Yin, Yangqiu Song, Ming Zhang2017 · 99 citations
5. FIMO: A Challenge Formal Dataset for Automated Theorem Proving ↗arXiv:2309.04295 · no summary yetChengwu Liu et al.2023 · 53 citations
6. DQ-LoRe: Dual Queries with Low Rank Approximation Re-ranking for In-Context LearningarXiv:2310.02954Jing Xiong, Zixuan Li, Chuanyang Zheng, et al.2023 · 27 citations
7. TRIGO: Benchmarking Formal Mathematical Proof Reduction for Generative Language Models ↗Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · no summary yetJing Xiong et al.2023 · 21 citations
8. Safe: Enhancing Mathematical Reasoning in Large Language Models via Retrospective Step-aware Formal Verification ↗arXiv:2506.04592 · no summary yetChengwu Liu et al.2025 · 17 citations
9. Pangu Ultra MoE: How to Train Your Big MoE on Ascend NPUsarXiv:2505.04519Yehui Tang, Yichun Yin, Yaoyuan Wang, et al.2025 · 8 citations
10. 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
11. MATCH: Modulating Attention via In-Context Retrieval for Long-Context TransformersAnnual Meeting of the Association for Computational Linguistics (ACL)Linrui Ma, Chun Hei Lo, Xinyu Wang, et al.2026

🤝 Co-authors here

12 of 29 co-authors with a page, the most papers together first.

• Lifeng Shang ×3
• Qun Liu ×2
• Wei Guo ×2
• Xiao Chen ×2
• Xin Jiang ×2
• Yaoyuan Wang ×2
• Yehui Tang ×2
• Can Chen
• Chengming Li
• Dong Li
• Fangcheng Liu
• Hang Zhou

Venues

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

• Conference on Empirical Methods in Natural Language Processing (EMNLP) ×3: EMNLP 2020
• Annual Meeting of the Association for Computational Linguistics (ACL): ACL 2026
• Findings of the Association for Computational Linguistics: Findings of the Association for Computational Linguistics 2020
• Findings of the Association for Computational Linguistics: EMNLP 2021

🔗 Cited by

Papers of our library cite their works 913 times (of the 149,439 papers whose bibliographies we read); 108 citing papers with a page here →.

Updated 1 November 2026

↑ back to top

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