--- title: "Shu Liu — author profile and papers — Elamur Research" description: "Shu Liu, Chinese Academy of Sciences: 4 papers on Elamur Research (2023–2025), h-index 5 — affiliations, topics, co-authors and who…" canonical: "https://research.elamur.ai/free/en/authors/shu-liu/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/shu-liu/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Shu Liu # Shu Liu Shu Liu has 4 papers in our library, in 2023–2025, most of them on [Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/), [Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/), [Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/); often with [Dacheng Li](https://md.research.elamur.ai/free/en/authors/dacheng-li/), [Ion Gabriel Stoica](https://md.research.elamur.ai/free/en/authors/ion-gabriel-stoica/), [Joseph E. Gonzalez](https://md.research.elamur.ai/free/en/authors/joseph-e-gonzalez/). 4 papers in our library · 2023–2025 [OpenAlex](https://openalex.org/A5023175954) · [ORCID](https://orcid.org/0000-0002-9812-7821) · [Wikidata](https://www.wikidata.org/wiki/Q91842421) · [**h-index 5**](https://openalex.org/authors/A5023175954) · [19 works](https://openalex.org/works?filter=author.id:A5023175954) · [67 citations](https://openalex.org/works?filter=author.id:A5023175954&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [📄 4 works in our library](https://md.research.elamur.ai/free/en/authors/shu-liu/#papers) · [🔗 Cited 187 times in our library](https://md.research.elamur.ai/free/en/authors/shu-liu/cited-by/) · [1 more in our library without a page yet](#also) · [⭐ Paper Stars ranking: #6944](https://md.research.elamur.ai/free/en/paper-stars/page/70/#star-6944) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Shu+Liu) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Shu+Liu) · [dblp](https://dblp.org/search?q=Shu+Liu) · [ACL Anthology](https://aclanthology.org/search/?q=Shu+Liu) ## Affiliations [🏛️**Chinese Academy of Sciences** Beijing · China 2021–2025 · 3 papers*current*](https://md.research.elamur.ai/free/en/government/chinese-academy-of-sciences/) · [🎓**Chongqing University of Arts and Sciences** Chongqing · China 2025–2026 · 2 papers*current*](https://ror.org/01rcvq140) · [🔬**National Astronomical Observatories** Beijing · China 2021–2025 · 3 papers](https://ror.org/058pyyv44) · [🎓**University of California, Berkeley** Berkeley · United States 2025 · 3 papers](https://md.research.elamur.ai/free/en/universities/university-of-california-berkeley/) · [🔬**Beijing Proteome Research Center** Beijing · China 2019–2022 · 3 papers](https://ror.org/05pp5b412) · [🔬**State Key Laboratory of Proteomics** Beijing 2019–2022 · 3 papers](https://ror.org/019bev041) 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 7 The field of science this work belongs to (OpenAlex fields).[Computer Science · 4 works](https://md.research.elamur.ai/free/en/shelves/computer-science/) · [Artificial Intelligence · 4 works](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · [Arts and Humanities · 3 works](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · [Language and Linguistics · 3 works](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · [Computer Vision and Pattern Recognition · 2 works](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · [Information Systems · 1 work](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · [Hardware and Architecture · 1 work](https://md.research.elamur.ai/free/en/shelves/computer-science/hardware-and-architecture/) 🧭 Topics 4 Research subjects and languages the work is about.[Language models & pretraining · 2 works](https://md.research.elamur.ai/free/en/topics/language-models/) · [Reasoning · 2 works](https://md.research.elamur.ai/free/en/topics/reasoning/) · [GPU & parallel computing · 1 work](https://md.research.elamur.ai/free/en/topics/gpu-parallel-computing/) · [Evaluation & benchmarks · 1 work](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) ## Works by year ### 2020s [**2025** 2 papers](https://md.research.elamur.ai/free/en/authors/shu-liu/year/2025/) · [**2024** 1 paper](https://md.research.elamur.ai/free/en/authors/shu-liu/year/2024/) · [**2023** 1 paper](https://md.research.elamur.ai/free/en/authors/shu-liu/year/2023/) ## Papers The most cited first (OpenAlex and our library). 1. **[LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters!](https://md.research.elamur.ai/free/en/papers/li-2025-llms-easily-learn-reason-demonstrations/)** *(2025 · 56 citations)* Large reasoning models solve hard problems through long chains of thought with self-checking and backtracking. 2. **[MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUs](https://md.research.elamur.ai/free/en/papers/cao-2024-moe-lightning-high-throughput-moe/)** *(2024 · 51 citations)* Mixture-of-Experts LLMs keep per-token compute low but need very large amounts of memory. The expert FFNs of Mixtral 8x22B alone take more than 256 GB. This paper targets offline, throughput-oriented batch inference, such as model evaluation, synthetic data generation and data wrangling, when the GPU cannot hold the… 3. **[The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks](https://md.research.elamur.ai/free/en/papers/cuadron-2025-danger-overthinking-examining-reasoning-action/)** *(2025 · 41 citations)* Large Reasoning Models (LRMs) such as o1, QwQ and DeepSeek-R1 perform well on static benchmarks. In agentic settings, however, they must keep choosing between acting in the environment and continuing to reason internally. The authors call this the Reasoning-Action Dilemma. They study "overthinking," where a model… 4. **[MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation](https://md.research.elamur.ai/free/en/papers/zeng-2023-mr-gsm8k-meta-reasoning-benchmark/)** *(2023 · 36 citations)* Math word-problem benchmarks such as GSM8K are close to saturation. Leading LLMs exceed 80% accuracy, scoring checks only the final answer, and test-set contamination may inflate results. The authors propose testing models as graders instead of solvers, a setup they call meta-reasoning. They built MR-GSM8K from three… ## All works 5 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. [**LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters! arXiv:2502.07374**Dacheng Li, Shiyi Cao, Tyler Griggs, et al.*2025 · 56 citations*](https://md.research.elamur.ai/free/en/papers/li-2025-llms-easily-learn-reason-demonstrations/) 2. [**MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUs ASPLOS 2025**Shiyi Cao, Shu Liu, Tyler Griggs, et al.*2024 · 51 citations*](https://md.research.elamur.ai/free/en/papers/cao-2024-moe-lightning-high-throughput-moe/) 3. [**The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks arXiv:2502.08235**Alejandro Cuadron, Dacheng Li, Wenjie Ma, et al.*2025 · 41 citations*](https://md.research.elamur.ai/free/en/papers/cuadron-2025-danger-overthinking-examining-reasoning-action/) 4. [**MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation arXiv:2312.17080**Zhongshen Zeng, Pengguang Chen, Shu Liu, et al.*2023 · 36 citations*](https://md.research.elamur.ai/free/en/papers/zeng-2023-mr-gsm8k-meta-reasoning-benchmark/) 5. **[Language Models Can Easily Learn to Reason from Demonstrations ↗](https://aclanthology.org/2025.findings-emnlp.866/) Findings of the Association for Computational Linguistics: EMNLP 2025 · no summary yet**Dacheng Li et al.*2025 · 14 citations* ## Co-authors here 12 of 18 co-authors with a page, the most papers together first. - [Dacheng Li](https://md.research.elamur.ai/free/en/authors/dacheng-li/) ×2 - [Ion Gabriel Stoica](https://md.research.elamur.ai/free/en/authors/ion-gabriel-stoica/) ×2 - [Joseph E. Gonzalez](https://md.research.elamur.ai/free/en/authors/joseph-e-gonzalez/) ×2 - [Matei Zaharia](https://md.research.elamur.ai/free/en/authors/matei-zaharia/) ×2 - [Shiyi Cao](https://md.research.elamur.ai/free/en/authors/shiyi-cao/) ×2 - [Tyler Griggs](https://md.research.elamur.ai/free/en/authors/tyler-griggs/) ×2 - [Alejandro Cuadron](https://md.research.elamur.ai/free/en/authors/alejandro-cuadron/) - [Eric Tang](https://md.research.elamur.ai/free/en/authors/eric-tang/) - [Haiyun Jiang](https://md.research.elamur.ai/free/en/authors/haiyun-jiang/) - [Huanzhi Mao](https://md.research.elamur.ai/free/en/authors/huanzhi-mao/) - [Jiaya Jia](https://md.research.elamur.ai/free/en/authors/jiaya-jia/) - [Shishir G. Patil](https://md.research.elamur.ai/free/en/authors/shishir-g-patil/) ## Venues Where their works here were published, the most works first, with the editions of those that have a page. - ASPLOS 2025 - Findings of the Association for Computational Linguistics: EMNLP 2025 ## Cited by Papers of our library cite their works 187 times (of the 149,439 papers whose bibliographies we read); [49 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/shu-liu/cited-by/). Updated 3 November 2026