--- title: "Yong Cheng — author profile and papers" description: "Yong Cheng, Shandong University: 2 papers on Elamur Research (2015–2019), h-index 34 — affiliations, topics, co-authors and who cites them in our library." canonical: "https://research.elamur.ai/free/en/authors/yong-cheng/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/yong-cheng/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Yong Cheng # Yong Cheng Yong Cheng works on neural machine translation, with papers here from 2015 to 2019. They ask how translation models should be trained: what objective best matches the way translations are judged, and how to keep a system stable when its input is perturbed. The methods come from model training itself, including risk-based objectives borrowed from statistical translation and adversarial example generation for Transformer systems. The most cited paper here is [Minimum Risk Training for Neural Machine Translation](https://md.research.elamur.ai/free/en/papers/shen-2015-minimum-risk-training-neural-machine/) (2015). It replaces standard maximum likelihood training with an objective that minimizes expected loss over the model's own candidate translations, measured with a sentence-level metric such as smoothed BLEU. Because summing over all possible outputs cannot be done, the method draws a small set of sampled translations, adds the reference, and renormalizes probabilities over that set. This lets an end-to-end system be tuned toward the quality measure used for evaluation rather than toward word-by-word accuracy on gold prefixes. The later paper, [Robust Neural Machine Translation with Doubly Adversarial Inputs](https://md.research.elamur.ai/free/en/papers/cheng-2019-robust-neural-machine-translation-doubly/) (2019), shifts from the training objective to robustness. It begins from the observation that small edits to a source sentence, such as replacing one pronoun in Chinese, can badly damage a Transformer's output. To address this, the paper trains word-level Transformer models on adversarial inputs generated with white-box access to the model. Cheng's papers here fall under [machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/), and the bylines list Google in Mountain View from 2019 to 2022. Co-authors include [Wolfgang Macherey](https://md.research.elamur.ai/free/en/authors/wolfgang-macherey/), [Maosong Sun](https://md.research.elamur.ai/free/en/authors/maosong-sun/), [Yang Liu](https://md.research.elamur.ai/free/en/authors/yang-liu/), [Hua Wu](https://md.research.elamur.ai/free/en/authors/hua-wu/) and [Zhongjun He](https://md.research.elamur.ai/free/en/authors/zhongjun-he/). Yong Cheng has 2 papers in our library, in 2015–2019, most of them on [Machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/); often with [Hua Wu](https://md.research.elamur.ai/free/en/authors/hua-wu/), [Maosong Sun](https://md.research.elamur.ai/free/en/authors/maosong-sun/), [Wei He](https://md.research.elamur.ai/free/en/authors/wei-he/). 2 papers in our library · 2015–2019 [OpenAlex](https://openalex.org/A5100674551) · [ORCID](https://orcid.org/0000-0002-6465-3022) · [**h-index 34**](https://openalex.org/authors/A5100674551) · [199 works](https://openalex.org/works?filter=author.id:A5100674551) · [4,220 citations](https://openalex.org/works?filter=author.id:A5100674551&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [📄 2 works in our library](https://md.research.elamur.ai/free/en/authors/yong-cheng/#papers) · [🔗 477 citations in our library](https://md.research.elamur.ai/free/en/authors/yong-cheng/cited-by/) · [4 more in our library without a page yet](#also) · [⭐ Paper Star rank #2140](https://md.research.elamur.ai/free/en/paper-stars/page/22/#star-2140) 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=Yong+Cheng) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Yong+Cheng) · [dblp](https://dblp.org/search?q=Yong+Cheng) · [ACL Anthology](https://aclanthology.org/search/?q=Yong+Cheng) ## Affiliations [🏛️**Chinese Academy of Sciences** Beijing · China 2005–2026 · 120 papers*current*](https://md.research.elamur.ai/free/en/government/chinese-academy-of-sciences/) · [🔬**Institute of Plasma Physics** Hefei · China 2010–2026 · 115 papers*current*](https://ror.org/033cbzv42) · [🔬**Hefei Institutes of Physical Science** Hefei · China 2019–2026 · 39 papers*current*](https://ror.org/046n57345) · [🎓**National University of Defense Technology** Changsha · China 2025–2026 · 3 papers*current*](https://md.research.elamur.ai/free/en/universities/national-university-of-defense-technology/) · [🎓**Shandong University** Jinan · China 2024–2026 · 3 papers*current*](https://md.research.elamur.ai/free/en/universities/shandong-university/) · [🎓**Nanjing University of Posts and Telecommunications** Nanjing · China 2026 · 1 paper*current*](https://ror.org/043bpky34) · [🎓**University of Science and Technology of China** Hefei · China 2017–2025 · 9 papers](https://md.research.elamur.ai/free/en/universities/university-of-science-and-technology-of-china/) · [🎓**Tsinghua University** Beijing · China 2014–2017 · 7 papers](https://md.research.elamur.ai/free/en/universities/tsinghua-university/) · [🔬**Institute of Energy, Hefei Comprehensive National Science Center** Hefei · China 2022–2023 · 5 papers](https://ror.org/036kqh429) · [🎓**Fudan University** Shanghai · China 2014–2017 · 5 papers](https://md.research.elamur.ai/free/en/universities/fudan-university/) From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 21 more institutions on single bylines are not listed. ## Topics [Machine translation (2)](https://md.research.elamur.ai/free/en/topics/machine-translation/) ## Works by year ### 2010s [**2019** 1 paper](https://md.research.elamur.ai/free/en/papers/cheng-2019-robust-neural-machine-translation-doubly/) · [**2015** 1 paper](https://md.research.elamur.ai/free/en/papers/shen-2015-minimum-risk-training-neural-machine/) ## Papers The most cited first (OpenAlex and our library). 1. **[Minimum Risk Training for Neural Machine Translation](https://md.research.elamur.ai/free/en/papers/shen-2015-minimum-risk-training-neural-machine/)** *(2015 · 572 citations)* Neural machine translation (NMT) is usually trained with maximum likelihood estimation (MLE). MLE optimizes word-level cross-entropy on gold prefixes, not the sentence-level metrics such as BLEU that measure translation quality, and it never shows the model its own outputs. The authors adapt minimum risk training… 2. **[Robust Neural Machine Translation with Doubly Adversarial Inputs](https://md.research.elamur.ai/free/en/papers/cheng-2019-robust-neural-machine-translation-doubly/)** *(2019 · 321 citations)* NMT systems can produce much worse output when the input changes only slightly. In one example, swapping "he" for "she" in a Chinese sentence derailed a Transformer translation. The authors address this weakness with a white-box adversarial training scheme for word-level Transformer models. ## Also in our library 4 more works of theirs are in our library without a page here yet; they will appear as we publish them. ## Co-authors here 6 co-authors with a page, the most papers together first. - [Hua Wu](https://md.research.elamur.ai/free/en/authors/hua-wu/) - [Maosong Sun](https://md.research.elamur.ai/free/en/authors/maosong-sun/) - [Wei He](https://md.research.elamur.ai/free/en/authors/wei-he/) - [Wolfgang Macherey](https://md.research.elamur.ai/free/en/authors/wolfgang-macherey/) - [Yang Liu](https://md.research.elamur.ai/free/en/authors/yang-liu/) - [Zhongjun He](https://md.research.elamur.ai/free/en/authors/zhongjun-he/) ## Cited by Papers of our library cite their works 477 times (of the 140,884 papers whose bibliographies we read); [45 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/yong-cheng/cited-by/).