🧑‍🔬 Yanghua Xiao

🏛️ Affiliations

🎓Fudan UniversityShanghai · China2026 · 14 paperscurrent

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine.

🧭 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. DeepSeek: Paradigm Shifts and Technical Evolution in Large AI Models (2025 · 41 citations)
   This survey, written by researchers at several Chinese universities and Swinburne University, explains why DeepSeek's V3 and R1 models drew so much attention and what they suggest about the direction of large AI models.
2. Instructions are all you need: Self-supervised Reinforcement Learning for Instruction Following (2026 · 3 citations)
   Language models often fail to satisfy instructions that contain several constraints at once. Existing RL methods for this problem depend on stronger models or human annotators for rewards, and multi-constraint tasks give sparse rewards. The authors propose a self-supervised RL framework in which reward signals come…
3. Why Did Apple Fall: Evaluating Curiosity in Large Language Models (2026 · 1 citation)
   Curiosity-inspired intrinsic rewards are increasingly used in RL training of LLMs, but it has not been checked whether these models actually behave in curious ways, or whether the reported gains simply come from amplified statistical signals such as entropy.
4. Don't Tell the Answer, Truly Guide the Reasoning During RL Rollouts (2026)
   GRPO-style reinforcement learning struggles when a problem exceeds what the model can currently solve. Every rollout fails, group advantages become zero, and the gradient disappears. Earlier fixes inject reference solutions, either whole or partial ("Answer-Hints"). The authors argue these fixes often clash with the…
5. HumanLLM: Benchmarking and Improving LLM Anthropomorphism via Human Cognitive Patterns (2026)
   LLM-based role-playing agents usually treat personality as a set of independent labels, each tied to a stereotyped behavior.
6. Memory-Guided Hard Data Augmentation for Multimodal Named Entity Recognition (2026)
   The paper targets data augmentation for Multimodal Named Entity Recognition (MNER), where images help disambiguate text.

🤝 Co-authors here

12 of 24 co-authors with a page, the most papers together first.

• Fei Yu ×2
• Jiansheng Wei ×2
• Jiaqing Liang ×2
• Sihang Jiang ×2
• Xiaojun Meng ×2
• Bo Xu
• Haofen Wang
• Haoyu Wang
• Huajun Chen
• Jen-tse Huang
• Jian Yang
• Jie Zeng

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 ×3: Findings of the Association for Computational Linguistics 2026 ×3
• Annual Meeting of the Association for Computational Linguistics (ACL) ×2: ACL 2026 ×2

🔗 Cited by

Papers of our library cite their works 3 times (of the 149,439 papers whose bibliographies we read); 2 citing papers with a page here →.

Updated 1 November 2026

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