Elamur Research⌕
☰
home / authors / Zhiheng Xi

Zhiheng Xi

Zhiheng Xi has 7 papers in our library, in 2023–2026, most of them on Language models & pretraining, Evaluation & benchmarks, Reasoning; often with Tao Gui, Xiaoran Fan, Enyu Zhou.

7 papers in our library · 2023–2026

OpenAlex · ORCID · h-index 7 · 112 works · 1,103 citations OpenAlex, 2026-10-14

📄 7 works in our library · 🔗 Cited 714 times in our library · 8 more in our library without a page yet · ⭐ Paper Stars ranking: #1922

The profile is the library's match of the name; two people with one name are not always told apart.

Elsewhere: Google Scholar · Semantic Scholar · dblp · ACL Anthology

 Affiliations

🎓Fudan UniversityShanghai · China2022–2027 · 82 papersEducation 2022–2027 · Computer ScienceEmploymentcurrent🏢TencentShenzhen · China2025–2026 · 3 paperscurrent

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.

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Arts and Humanities · 7 worksComputer Science · 7 worksArtificial Intelligence · 7 worksLanguage and Linguistics · 7 worksComputer Vision and Pattern Recognition · 1 workInformation Systems · 2 works

🧭 Topics 3Research subjects and languages the work is about.Language models & pretraining · 3 worksEvaluation & benchmarks · 2 worksReasoning · 1 work

💡 Concepts 1Key ideas and methods explained on our concept pages.Large language models · 1 work

 Works by year

2020s

20263 papers20251 paper20242 papers20231 paper

 Papers

The most cited first (OpenAlex and our library).

  1. The Rise and Potential of Large Language Model Based Agents: A Survey (2023 · 1,272 citations)
    This 2023 survey from the Fudan NLP Group maps the young field of agents built on large language models. The authors argue that earlier agent research mostly tuned algorithms for narrow skills, such as symbolic reasoning or board games like Go and Chess. What the field lacked was a general base model that agents…
  2. LoRAMoE: Alleviating World Knowledge Forgetting in Large Language Models via MoE-Style Plugin (2024 · 93 citations)
    The paper examines a trade-off in supervised fine-tuning (SFT) of LLMs. Adding much more instruction data keeps improving downstream tasks, but it also erodes the factual knowledge the model gained in pre-training. The authors build a 5-million-sample instruction set covering seven task categories and fine-tune…
  3. RMB: Comprehensively Benchmarking Reward Models in LLM Alignment (2024 · 30 citations)
    Reward models (RMs) steer LLM alignment, but existing evaluations such as RewardBench often fail to predict how well an RM actually guides alignment.
  4. ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool Use (2025 · 15 citations)
    Benchmarks for LLM tool use rarely test true multi-hop behaviour. Most build datasets tool-first: they gather tools and then synthesize queries for them. Such tools often do not depend on one another, and answers cannot be checked reliably, so evaluation falls back on model-based judging, which carries bias. ToolHop…
  5. Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build Environments (2026 · 6 citations)
    Training LLMs to call tools with reinforcement learning is difficult for two reasons. Online APIs make training environments unstable and costly to maintain, and judging trajectories usually requires a strong LLM as evaluator, which brings bias. The authors propose FTRL, which removes both problems. An automated…
  6. Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment (2026 · 4 citations)
    Distilling reasoning from a teacher LLM into a student through SFT on long chain-of-thought (CoT) traces does not reliably improve with teacher strength.
  7. Enhancing LLM-based Search Agents via Contribution Weighted Group Relative Policy Optimization (2026)
    Search agents built on LLMs learn to query external engines during multi-round reasoning, but reinforcement learning for such agents faces a dilemma.

All works

15 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. The Rise and Potential of Large Language Model Based Agents: A SurveyScience China Information SciencesZhiheng Xi, Wenxiang Chen, Xin Guo, et al.2023 · 1,272 citations
  2. LoRAMoE: Alleviating World Knowledge Forgetting in Large Language Models via MoE-Style PluginAnnual Meeting of the Association for Computational Linguistics (ACL)Shihan Dou, Enyu Zhou, Yan Liu, et al.2024 · 93 citations
  3. Secrets of RLHF in Large Language Models Part II: Reward Modeling ↗arXiv:2401.06080 · no summary yetBinghai Wang et al.2024 · 72 citations
  4. Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learning ↗arXiv:2402.05808 · no summary yetZhiheng Xi et al.2024 · 36 citations
  5. AgentGym: Evolving Large Language Model-based Agents across Diverse Environments ↗arXiv:2406.04151 · no summary yetZhiheng Xi et al.2024 · 33 citations
  6. StepCoder: Improving Code Generation with Reinforcement Learning from Compiler Feedback ↗Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · no summary yetShihan Dou et al.2024 · 32 citations
  7. RMB: Comprehensively Benchmarking Reward Models in LLM AlignmentarXiv:2410.09893Enyu Zhou, Guodong Zheng, Binghai Wang, et al.2024 · 30 citations
  8. ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool UseAnnual Meeting of the Association for Computational Linguistics (ACL)Junjie Ye, Zhengyin Du, Xuesong Yao, et al.2025 · 15 citations
  9. Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build EnvironmentsFindings of the Association for Computational LinguisticsJunjie Ye, Changhao Jiang, Zhengyin Du, et al.2026 · 6 citations
  10. Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative AlignmentAnnual Meeting of the Association for Computational Linguistics (ACL)Yuming Yang, Mingyoung Lai, Wanxu Zhao, et al.2026 · 4 citations
  11. Enhancing LLM-based Search Agents via Contribution Weighted Group Relative Policy OptimizationAnnual Meeting of the Association for Computational Linguistics (ACL)Junzhe Wang, Zhiheng Xi, Yajie Yang, et al.2026
  12. AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments ↗Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · no summary yetZhiheng Xi et al.2026
  13. ABC-Bench: Benchmarking Agentic Backend Coding in Real-World Development ↗Findings of the Association for Computational Linguistics: ACL 2026 · no summary yetJie Yang et al.2026
  14. AgentV-RL: Scaling Reward Modeling with Agentic Verifier ↗Findings of the Association for Computational Linguistics: ACL 2026 · no summary yetJiazheng Zhang et al.2026
  15. VRPO: Rethinking Value Modeling for Robust RL under Noisy Supervision in LLM Post-Training ↗Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · no summary yetDingwei Zhu et al.2026

 Co-authors here

12 of 34 co-authors with a page, the most papers together first.

Venues

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

 Cited by

Papers of our library cite their works 714 times (of the 149,439 papers whose bibliographies we read); 145 citing papers with a page here →.

Updated

↑ back to top

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