Elamur Research⌕
☰
home / authors / Songyang Gao

Songyang Gao

Songyang Gao has 5 papers in our library, in 2023–2025, most of them on Language models & pretraining, Reasoning; often with Shihan Dou, Limao Xiong, Wei Shen.

5 papers in our library · 2023–2025

OpenAlex · ORCID · h-index 8 · 45 works · 200 citations OpenAlex, 2026-10-14

📄 5 works in our library · 🔗 Cited 295 times in our library · 2 more in our library without a page yet · ⭐ Paper Stars ranking: #5102

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

🔬Shanghai Artificial Intelligence LaboratoryShanghai · China2024–2026 · 6 paperscurrent🏥Daqing Oilfield General HospitalDaqing · China2026 · 1 papercurrent🔬Research Institute of Petroleum Exploration and DevelopmentBeijing · China2026 · 1 papercurrent🎓The Hong Kong University of Science and Technology (Guangzhou)Guangzhou · China2026 · 1 papercurrent🎓Fudan UniversityShanghai · China2022–2025 · 20 papers🎓Dalian University of TechnologyDalian · China2022–2023 · 2 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 9 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 9The field of science this work belongs to (OpenAlex fields).Computer Science · 5 worksArtificial Intelligence · 5 worksArts and Humanities · 4 worksLanguage and Linguistics · 4 worksEngineering · 1 workComputer Vision and Pattern Recognition · 1 workControl and Systems Engineering · 1 workApplied Mathematics · 1 workMathematics · 1 work

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 4 worksReasoning · 1 work

 Works by year

2020s

20253 papers20241 paper20231 paper

 Papers

The most cited first (OpenAlex and our library).

  1. Secrets of RLHF in Large Language Models Part I: PPO (2023 · 135 citations)
    RLHF is the main way to make LLMs helpful, honest and harmless. Its PPO stage, though, is unstable, sensitive to hyperparameters and costly to debug, and few open implementations existed. This report from the Fudan NLP Group and ByteDance breaks the RLHF pipeline into its stages: SFT, reward model training, and PPO…
  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. ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world Scenarios (2025 · 35 citations)
    Most benchmarks for tool use by large language models only check whether a model picks the expected API. That approach works only when correct answers can be fixed in advance, and it reveals little about the skills behind successful tool use. ToolEyes addresses both gaps. The authors define seven realistic scenarios…
  4. CompassVerifier: A Unified and Robust Verifier for LLMs Evaluation and Outcome Reward (2025 · 16 citations)
    Checking whether an LLM's free-form output matches a reference answer is central to benchmarking and to reward signals in RL.
  5. Intern-S1-MO: Long-horizon Reasoning Agent for Olympiad?Level Mathematical Problem Solving (2025 · 1 citation)
    Even strong Large Reasoning Models (LRMs) struggle on IMO-level problems. The exploration these problems need outgrows a single 64K–128K context window, and earlier workarounds were mostly prompt-only pipelines on proprietary models. The authors present Intern-S1-MO, an agent framework that splits problem solving…

All works

7 works of theirs in our library, the most cited first; 5 with a page here, the others link their original until we summarise them.

  1. Secrets of RLHF in Large Language Models Part I: PPOarXiv:2307.04964Rui Zheng, Shihan Dou, Songyang Gao, et al.2023 · 135 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. ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world ScenariosInternational Conference on Computational Linguistics (COLING)Junjie Ye, Guanyu Li, SongYang Gao, et al.2025 · 35 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. CompassVerifier: A Unified and Robust Verifier for LLMs Evaluation and Outcome RewardConference on Empirical Methods in Natural Language Processing (EMNLP)Shudong Liu, Hongwei Liu, Junnan Liu, et al.2025 · 16 citations
  7. Intern-S1-MO: Long-horizon Reasoning Agent for Olympiad?Level Mathematical Problem SolvingarXiv:2512.10739Yuzhe Gu, Songyang Gao, Zijian Wu, et al.2025 · 1 citation

 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 295 times (of the 149,439 papers whose bibliographies we read); 59 citing papers with a page here →.

Updated

↑ back to top

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