🧑‍🔬 Bohan Li

🏛️ Affiliations

🎓Nanjing University of Aeronautics and AstronauticsNanjing · China2009–2026 · 119 paperscurrent🏛️Ministry of Industry and Information TechnologyBeijing · China2019–2026 · 33 paperscurrent🎓Shanghai Jiao Tong UniversityShanghai · China2023–2026 · 7 paperscurrent🎓Huazhong University of Science and TechnologyWuhan · China2016–2026 · 2 paperscurrent🎓Stanford UniversityStanford · United States2024–2026 · 2 paperscurrent🎓University of BathBath · United Kingdom2025–2026 · 2 paperscurrent🎓University of California, Santa BarbaraSanta Barbara · United States2016–2020 · 4 papers🎓Eastern Institute of Technology, NingboNingbo · China2023–2024 · 3 papers🎓Henan Polytechnic UniversityJiaozuo · China2023–2024 · 3 papers🎓Northwestern Polytechnical UniversityXi'an · China2023–2025 · 2 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 36 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. On the Sentence Embeddings from Pre-trained Language Models (2020 · 833 citations)
   Sentence vectors taken directly from BERT, without fine-tuning, are poor at semantic textual similarity. They do worse than averaged GloVe vectors. The authors ask whether BERT lacks semantic knowledge or whether that knowledge is simply hard to reach with cosine similarity.
2. MathAgent: Adversarial Evolution of Constraint Graphs for Mathematical Reasoning Data Synthesis (2026 · 3 citations)
   Synthetic data for mathematical reasoning is often built either by mutating human-curated seed problems, which limits diversity to the seeds' scope, or by prompting models from scratch, which tends toward repetitive, shallow problems (mode collapse).
3. BSCodec: A Band-Split Neural Codec for High-Quality Universal Audio Reconstruction (2026 · 1 citation)
   Neural audio codecs tuned for speech tend to degrade on music and environmental sound, whose energy is spread differently across the spectrum.

All works

4 works of theirs in our library, the most cited first; 3 with a page here, the others link their original until we summarise them.

1. On the Sentence Embeddings from Pre-trained Language ModelsConference on Empirical Methods in Natural Language Processing (EMNLP)Bohan Li, Hao Zhou, Junxian He, et al.2020 · 833 citations
2. An Adversarial Approach to High-Quality, Sentiment-Controlled Neural Dialogue Generation ↗arXiv (Cornell University) · no summary yetXiang Kong et al.2019 · 24 citations
3. MathAgent: Adversarial Evolution of Constraint Graphs for Mathematical Reasoning Data SynthesisFindings of the Association for Computational LinguisticsZixiong Yu, Jun Rao, Guhan Chen, et al.2026 · 3 citations
4. BSCodec: A Band-Split Neural Codec for High-Quality Universal Audio ReconstructionFindings of the Association for Computational LinguisticsHaoran Wang, Jiatong Shi, Jinchuan Tian, et al.2026 · 1 citation

🤝 Co-authors here

12 of 15 co-authors with a page, the most papers together first.

• Hao Zhou
• Haoran Wang
• Jiansheng Wei
• Jiatong Shi
• Jinchuan Tian
• Jun Rao
• Junxian He
• Kai Yu
• Lei Li
• Min Zhang
• Mingxuan Wang
• Shinji Watanabe

Venues

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

• Findings of the Association for Computational Linguistics ×2: Findings of the Association for Computational Linguistics 2026 ×2
• Conference on Empirical Methods in Natural Language Processing (EMNLP): EMNLP 2020

🔗 Cited by

Papers of our library cite their works 269 times (of the 149,439 papers whose bibliographies we read); 24 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/bohan-li/