Elamur Research⌕
☰
home / authors / Xinyang Geng

Xinyang Geng

Xinyang Geng has 3 papers in our library, in 2017–2024, most of them on Language models & pretraining, Reasoning; often with Pieter Abbeel, Adam Fisch, Alekh Agarwal.

3 papers in our library · 2017–2024

OpenAlex · ORCID · h-index 13 · 27 works · 1,472 citations OpenAlex, 2026-10-14

📄 3 works in our library · 🔗 Cited 345 times in our library · 3 more in our library without a page yet · ⭐ Paper Stars ranking: #4479

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

🎓University of California, BerkeleyBerkeley · United States2017–2024 · 19 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 4The field of science this work belongs to (OpenAlex fields).Artificial Intelligence · 3 worksComputer Science · 3 worksArts and Humanities · 2 worksLanguage and Linguistics · 2 works

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

 Works by year

2020s

20241 paper20231 paper

2010s

20171 paper

 Papers

The most cited first (OpenAlex and our library).

  1. The False Promise of Imitating Proprietary LLMs (2023 · 182 citations)
    Many open-source chat models, including Alpaca, Vicuna and Koala, are built by fine-tuning a weaker model on ChatGPT outputs.
  2. Automatic Goal Generation for Reinforcement Learning Agents (2017 · 181 citations)
    Standard reinforcement learning trains an agent to optimize a single reward, which works poorly when a robot must handle many related tasks, such as reaching arbitrary positions.
  3. Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoning (2024 · 93 citations)
    Process reward models (PRMs) score every step of a reasoning trace instead of only the final answer. Step labels written by humans are expensive, though, and earlier automatically labelled PRMs gave only marginal gains over outcome reward models (ORMs). This work asks what a step-level reward should actually measure.…

All works

6 works of theirs in our library, the most cited first; 3 with a page here, the others link their original until we summarise them.

  1. The False Promise of Imitating Proprietary LLMsarXiv:2305.15717Arnav Gudibande, Eric Wallace, Charlie Snell, et al.2023 · 182 citations
  2. Automatic Goal Generation for Reinforcement Learning AgentsarXiv:1705.06366Carlos Florensa, David Held, Xinyang Geng, et al.2017 · 181 citations
  3. Sequential Modeling Enables Scalable Learning for Large Vision Models ↗IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · no summary yetYutong Bai et al.2024 · 135 citations
  4. Rewarding Progress: Scaling Automated Process Verifiers for LLM ReasoningarXiv:2410.08146Amrith Setlur, Chirag Nagpal, Adam Fisch, et al.2024 · 93 citations
  5. RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold ↗arXiv:2406.14532 · no summary yetAmrith Setlur et al.2024 · 56 citations
  6. Rewriting History with Inverse RL: Hindsight Inference for Policy Improvement ↗arXiv (Cornell University) · no summary yetBenjamin Eysenbach et al.2020 · 38 citations

 Co-authors here

12 of 13 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 345 times (of the 149,439 papers whose bibliographies we read); 83 citing papers with a page here →.

Updated

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