Fuzheng Zhang
Fuzheng Zhang has 4 papers in our library, in 2021–2026, most of them on Reasoning, Question answering, Information retrieval & search; often with Bo Wang, Defu Lian, Guorui Zhou.
OpenAlex · ORCID · Wikidata · h-index 3 · 10 works · 32 citations OpenAlex, 2026-10-14
📄 4 works in our library · 🔗 Cited 225 times in our library · 1 more in our library without a page yet · ⭐ Paper Stars ranking: #3165
Affiliations
Fields, topics and concepts
📚 Shelves 8The field of science this work belongs to (OpenAlex fields).Computer Science · 4 worksArts and Humanities · 3 worksArtificial Intelligence · 4 worksLanguage and Linguistics · 3 worksInformation Systems · 2 worksDecision Sciences · 1 workComputer Vision and Pattern Recognition · 1 workManagement Science and Operations Research · 1 work
🧭 Topics 3Research subjects and languages the work is about.Reasoning · 2 worksInformation retrieval & search · 1 workQuestion answering · 1 work
Works by year
2020s
Papers
- ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfer (2021 · 647 citations)
Sentence vectors taken directly from BERT, whether the [CLS] output or averaged token embeddings, behave poorly on semantic textual similarity (STS). - DualRAG: A Dual-Process Approach to Integrate Reasoning and Retrieval for Multi-Hop Question Answering (2025 · 20 citations)
Multi-hop question answering requires a model to chain several reasoning steps, and the facts it needs change as the reasoning unfolds. - What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought Reasoning (2025 · 11 citations)
Long chain-of-thought (LCoT) reasoning lets LLMs solve hard problems, but it is unclear which properties of a reasoning trace indicate that the final answer will be right. - CE-GPPO: Coordinating Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning (2026 · 7 citations)
The paper tackles unstable policy entropy when large language models are trained with RL for reasoning. PPO-style methods, including GRPO and DAPO, clip the importance sampling ratio, and this silently drops gradients from low-probability tokens. The authors split tokens by advantage sign and probability. They argue…
All works
- ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation TransferAnnual Meeting of the Association for Computational Linguistics (ACL)Yuanmeng Yan, Rumei Li, Sirui Wang, et al.2021 · 647 citations
- Leanabell-Prover: Posttraining Scaling in Formal Reasoning ↗arXiv:2504.06122 · no summary yetJingyuan Zhang et al.2025 · 21 citations
- DualRAG: A Dual-Process Approach to Integrate Reasoning and Retrieval for Multi-Hop Question AnsweringAnnual Meeting of the Association for Computational Linguistics (ACL)Rong Cheng, Jinyi Liu, Yan Zheng, et al.2025 · 20 citations
- What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought ReasoningConference on Empirical Methods in Natural Language Processing (EMNLP)Gangwei Jiang, Yahui Liu, Zhaoyi Li, et al.2025 · 11 citations
- CE-GPPO: Coordinating Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement LearningAnnual Meeting of the Association for Computational Linguistics (ACL)Zhenpeng Su, Leiyu Pan, Minxuan Lv, et al.2026 · 7 citations
Co-authors here
- Bo Wang
- Defu Lian
- Guorui Zhou
- Kun Gai
- Leiyu Pan
- Linqi Song
- Minxuan Lv
- Rumei Li
- Sirui Wang
- Wei Bi
- Wei Wu
- Weiran Xu
Venues
- Annual Meeting of the Association for Computational Linguistics (ACL) ×3: ACL 2026, ACL 2025, ACL 2021
- Conference on Empirical Methods in Natural Language Processing (EMNLP): EMNLP 2025
Cited by
Updated
This page in Markdown → https://md.research.elamur.ai/free/en/authors/fuzheng-zhang/