🧑‍🔬 Bo An

🏛️ Affiliations

🎓Nanyang Technological UniversitySingapore · Singapore2013–2026 · 242 papersEmployment 2013–2018 · Nanyang Assistant Professor, Computer EngineeringEmployment since 2018 · Associate Professor, Computer Science and Engineeringcurrent🎓University of Southern CaliforniaLos Angeles · United States2010–2021 · 21 papersEmployment 2010–2012 · Postdoctoral Researcher, Computer Science🎓University of Massachusetts AmherstAmherst Center · United States2006–2011 · 19 papersEducation 2006–2010 · PhD, Computer Science🎓Peking UniversityBeijing · China2015–2020 · 12 papers🎓Shijiazhuang University of Applied TechnologyShijiazhuang · China2024–2025 · 4 papers🎓Heilongjiang Institute of TechnologyHarbin · China2001–2016 · 4 papers🔬Institute of Computing TechnologyBeijing · China2012–2013 · 3 papers🎓Hebei Chemical and Pharmaceutical CollegeShijiazhuang · China2022–2023 · 2 papers🏛️Chinese Academy of SciencesBeijing · China2012–2013 · 2 papersEmployment 2012–2013 · Associate Professor, Institute of Computing Technology🎓Chongqing UniversityChongqing · China1999–2006 · 2 papersEducation 1999–2003 · Bachlor, Computer ScienceEducation 2003–2006 · Master, Computer Science

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 19 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. Group-in-Group Policy Optimization for LLM Agent Training (2025 · 69 citations)
   Group-based RL methods such as GRPO work well for single-turn LLM tasks like math. They struggle with multi-turn agents, where episodes can run for dozens of steps and rewards often come only at the end, so individual actions are hard to credit. The authors propose GiGPO (Group-in-Group Policy Optimization), which…
2. Q*: Improving Multi-step Reasoning for LLMs with Deliberative Planning (2024 · 53 citations)
   Because LLMs generate text token by token, an early mistake in a multi-step solution tends to carry through and spoil the final answer.
3. LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification (2026 · 5 citations)
   Speculative decoding (SD) speeds up LLM inference without changing outputs. However, leading SD drafters such as EAGLE are trained on short sequences (about 2k–4k tokens) and break down when contexts get very long. The authors identify three causes:
4. SeDev: Structured Semantic Exploration for LLM-Driven Code Generation (2026)
   LLM-based multi-agent systems still struggle to build complete, multi-module software. Feedback-refinement pipelines follow a single trajectory and tend to settle in local optima. MCTS-style search explores more alternatives but becomes very costly. SeDev addresses this by making exploration explicit and cheap.

All works

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

1. POI2Vec: Geographical Latent Representation for Predicting Future Visitors ↗Proceedings of the AAAI Conference on Artificial Intelligence · no summary yetShanshan Feng et al.2017 · 225 citations
2. Deploying PAWS: Field Optimization of the Protection Assistant for Wildlife Security ↗Proceedings of the AAAI Conference on Artificial Intelligence · no summary yetFei Fang et al.2016 · 186 citations
3. Group-in-Group Policy Optimization for LLM Agent TrainingConference on Neural Information Processing Systems (NeurIPS)Lang Feng, Zhenghai Xue, Tingcong Liu, et al.2025 · 69 citations
4. Personalized Adaptive Meta Learning for Cold-start User Preference Prediction ↗Proceedings of the AAAI Conference on Artificial Intelligence · no summary yetRunsheng Yu et al.2021 · 59 citations
5. Q*: Improving Multi-step Reasoning for LLMs with Deliberative PlanningarXiv:2406.14283Chaojie Wang, Yanchen Deng, Zhiyi Lyu, et al.2024 · 53 citations
6. LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and VerificationAnnual Meeting of the Association for Computational Linguistics (ACL)Penghui Yang, Cunxiao Du, Fengzhuo Zhang, et al.2026 · 5 citations
7. SeDev: Structured Semantic Exploration for LLM-Driven Code GenerationAnnual Meeting of the Association for Computational Linguistics (ACL)Ronghui Yang, Jie Liu, Jiajie Zeng, et al.2026

🤝 Co-authors here

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

• Chao Du
• Chaojie Wang
• Cunxiao Du
• Haonan Wang
• Jie Liu
• Jiexin Wang
• Jujie He
• Liang Zeng
• Mengchen Zhao
• Shuicheng Yan
• Tianyu Pang
• Yi Cai

Venues

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

• Proceedings of the AAAI Conference on Artificial Intelligence ×3
• Annual Meeting of the Association for Computational Linguistics (ACL) ×2: ACL 2026 ×2
• Conference on Neural Information Processing Systems (NeurIPS): NeurIPS 2025

🔗 Cited by

Papers of our library cite their works 116 times (of the 149,439 papers whose bibliographies we read); 35 citing papers with a page here →.

Updated 2 November 2026

↑ back to top

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