---
title: "Sebastian Gehrmann — author profile and papers"
description: "Sebastian Gehrmann, Bloomberg (United States): 8 papers on Elamur Research (2018–2023), h-index 32 — affiliations, topics, co-authors and who cites them…"
canonical: "https://research.elamur.ai/free/en/authors/sebastian-gehrmann/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/sebastian-gehrmann/)

[home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Sebastian Gehrmann

# Sebastian Gehrmann

8 papers on this site · 2018–2023

current affiliation [Bloomberg (United States)](https://ror.org/02rdpzb15), New York, United States; [Harvard University](https://ror.org/03vek6s52), Cambridge, United States; [Johns Hopkins University](https://ror.org/00za53h95), Baltimore, United States; [Charles University](https://ror.org/024d6js02), Prague, Czechia; [Virginia Tech](https://ror.org/02smfhw86), Blacksburg, United States

Sebastian Gehrmann has 8 papers on this site, in 2018–2023, most of them on [Machine translation](https://md.research.elamur.ai/free/en/topics/machine-translation/), [Multilingual & cross-lingual NLP](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/), [Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/); often with [Ankur Parikh](https://md.research.elamur.ai/free/en/authors/ankur-parikh/), [Hyung Won Chung](https://md.research.elamur.ai/free/en/authors/hyung-won-chung/), [Alexander M. Rush](https://md.research.elamur.ai/free/en/authors/alexander-m-rush/).

[OpenAlex](https://openalex.org/A5026789962) · [ORCID](https://orcid.org/0000-0002-8257-9516) · [**h-index 32**](https://openalex.org/authors/A5026789962) · [87 works](https://openalex.org/works?filter=author.id:A5026789962) · [8 086 citations](https://openalex.org/works?filter=author.id:A5026789962&sort=cited_by_count:desc) OpenAlex, 2026-10-14

[8 works on this site](https://md.research.elamur.ai/free/en/authors/sebastian-gehrmann/#papers) · [4 294 citations in our library](https://md.research.elamur.ai/free/en/authors/sebastian-gehrmann/cited-by/) · 1 more work in our library

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=Sebastian+Gehrmann) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Sebastian+Gehrmann) · [dblp](https://dblp.org/search?q=Sebastian+Gehrmann) · [ACL Anthology](https://aclanthology.org/search/?q=Sebastian+Gehrmann)

## Affiliations

- [Harvard University](https://ror.org/03vek6s52) · Cambridge, United States —current · 2017–2026 on 30 papers
- [Bloomberg (United States)](https://ror.org/02rdpzb15) · New York, United States —current · 2022–2026 on 6 papers
- [Johns Hopkins University](https://ror.org/00za53h95) · Baltimore, United States —current · 2024–2026 on 2 papers
- [Charles University](https://ror.org/024d6js02) · Prague, Czechia —current · 2026 on 1 paper
- [Virginia Tech](https://ror.org/02smfhw86) · Blacksburg, United States —current · 2026 on 1 paper
- [Google (United States)](https://ror.org/00njsd438) · Mountain View, United States —2020–2023 on 33 papers
- [Stanford University](https://ror.org/00f54p054) · Stanford, United States —2022–2023 on 2 papers
- [University of Washington](https://ror.org/00cvxb145) · Seattle, United States —2021–2022 on 2 papers
- [Massachusetts Institute of Technology](https://ror.org/042nb2s44) · Cambridge, United States —2017–2020 on 4 papers
- [Bielefeld University](https://ror.org/02hpadn98) · Bielefeld, Germany —2018–2020 on 3 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 15 more institutions on single bylines are not listed.

## Topics

[Machine translation 2](https://md.research.elamur.ai/free/en/topics/machine-translation/) · [Multilingual & cross-lingual NLP 2](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/) · [Evaluation & benchmarks 2](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · [Language models & pretraining 1](https://md.research.elamur.ai/free/en/topics/language-models/) · [Summarization 1](https://md.research.elamur.ai/free/en/topics/summarization/) · [Text generation 1](https://md.research.elamur.ai/free/en/topics/text-generation/) · [Reasoning 1](https://md.research.elamur.ai/free/en/topics/reasoning/)

## Papers

The most cited first (OpenAlex and our library).

1. **[Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://md.research.elamur.ai/free/en/papers/suzgun-2023-challenging-big-bench-tasks-whether/)** *(2023 · 1 094 citations)* BIG-Bench was built to test large language models on tasks thought to be too hard for them. Yet the strongest model in the original evaluation, PaLM 540B, already beat the average human rater on most of its tasks. The authors ask whether the remaining tasks are truly out of reach or whether the original few-shot…
2. **[Bottom-Up Abstractive Summarization](https://md.research.elamur.ai/free/en/papers/gehrmann-2018-bottom-up-abstractive-summarization/)** *(2018 · 939 citations)* Neural abstractive summarizers write fluent text, but they often choose the wrong content. Pointer-generator models in particular tend to copy whole source sentences. Gehrmann, Deng and Rush address this with a two-stage design inspired by bottom-up attention in image captioning.
3. **[GLTR: Statistical Detection and Visualization of Generated Text](https://md.research.elamur.ai/free/en/papers/gehrmann-2019-gltr-statistical-detection-visualization-generated/)** *(2019 · 691 citations)* Large language models such as GPT-2 can produce text that ordinary readers struggle to tell apart from human writing, which opens the door to fake reviews, comments and news.
4. **[BloombergGPT: A Large Language Model for Finance](https://md.research.elamur.ai/free/en/papers/wu-2023-bloomberggpt-large-language-model-finance/)** *(2023 · 608 citations)* General-purpose LLMs dominate the field, but finance has its own vocabulary and tasks, such as sentiment analysis, NER, news classification and QA.
5. **[ToTTo: A Controlled Table-To-Text Generation Dataset](https://md.research.elamur.ai/free/en/papers/parikh-2020-totto-controlled-table-text-generation/)** *(2020 · 479 citations)* ToTTo is an open-domain English dataset for table-to-text generation from Google Research. It targets a known weakness of neural generators: they often hallucinate content that the source data does not support. Each input is a Wikipedia table with its metadata and a set of highlighted cells, and the model must write…
6. **[Learning Compact Metrics for MT](https://md.research.elamur.ai/free/en/papers/pu-2021-learning-compact-metrics-mt/)** *(2021 · 130 citations)* Learned MT evaluation metrics like COMET and BLEURT depend on large multilingual pre-trained encoders. Big models are expensive to run next to an MT system, but small ones handle many languages poorly. Using RemBERT and the WMT Metrics 2020 benchmark, the authors first show that capacity limits cross-lingual…
7. **[NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation](https://md.research.elamur.ai/free/en/papers/dhole-2023-nl-augmenter-framework-task-sensitive/)** *(2023 · 84 citations)* Data augmentation is used both to stress-test NLP models and to make their training data more diverse. Most existing perturbation tools offer a narrow, largely English set of edits that are rarely matched to a specific task. NL-Augmenter addresses this with an open, community-built Python library. It grew out of the…
8. **[Learning to Evaluate Translation Beyond English: BLEURT Submissions to the WMT Metrics 2020 Shared Task](https://md.research.elamur.ai/free/en/papers/sellam-2020-learning-evaluate-translation-beyond-english/)** *(2020 · 55 citations)* Learned metrics for machine translation evaluation mostly target English output, while the WMT Metrics 2020 Shared Task also judges translation into other languages, some of which have no human ratings.

## Also in our library

1 more work of theirs is in our library without a page here yet; it will appear as we publish them.

## Co-authors here

12 of 30 co-authors with a page, the most papers together first.

- [Ankur Parikh](https://md.research.elamur.ai/free/en/authors/ankur-parikh/) ×3
- [Hyung Won Chung](https://md.research.elamur.ai/free/en/authors/hyung-won-chung/) ×3
- [Alexander M. Rush](https://md.research.elamur.ai/free/en/authors/alexander-m-rush/) ×2
- [Amy Pu](https://md.research.elamur.ai/free/en/authors/amy-pu/) ×2
- [Dipanjan Das](https://md.research.elamur.ai/free/en/authors/dipanjan-das/) ×2
- [Thibault Sellam](https://md.research.elamur.ai/free/en/authors/thibault-sellam/) ×2
- [Aakanksha Chowdhery](https://md.research.elamur.ai/free/en/authors/aakanksha-chowdhery/)
- [Bhuwan Dhingra](https://md.research.elamur.ai/free/en/authors/bhuwan-dhingra/)
- [Denny Zhou](https://md.research.elamur.ai/free/en/authors/denny-zhou/)
- [Diyi Yang](https://md.research.elamur.ai/free/en/authors/diyi-yang/)
- [Ed Chi](https://md.research.elamur.ai/free/en/authors/ed-chi/)
- [Jascha Sohl-Dickstein](https://md.research.elamur.ai/free/en/authors/jascha-sohl-dickstein/)

## Cited by

Papers of our library cite their works 4 294 times (of the 140 884 papers whose bibliographies we read); [349 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/sebastian-gehrmann/cited-by/).