Awesome Speech Audio
Papers
Topics
Trending
Leaderboards
Authors
Datasets
Learn
Ask AI
Map
Tools
Packs
News
Videos
All sections
Browse
Papers
The full index, filterable and sortable.
Topics
The same papers grouped by subject tag.
Trending
What moved this week, and by how much.
Authors
Who publishes here, and who they publish with.
Map
The collection laid out by embedding similarity.
Compare
Leaderboards
Benchmark tables, with the paper behind each number.
Datasets
The datasets these papers train and evaluate on.
Tools
Code and libraries released alongside the papers.
Read
Learn
Ordered routes from background reading to current work.
Ask AI
Ask a question and get answers cited to these papers.
Videos
The most-watched talks and lectures in this field.
Packs
Short curated sets built around one question.
Follow
News
Press and coverage tied back to the papers.
Blogs
Author and lab write-ups of their own work.
Newsletter
Email digest of what changed, on a schedule.
Research Radar
Paste an abstract, get matches across every collection.
Yours
Saved
Papers you bookmarked in this browser.
+ Add Paper
☾
☀
← authors
·
overview
Loading author…
Ask AI
Mara Finkelstein — most-cited papers & profile · Speech Audio
← authors
·
overview
Mara Finkelstein
10
papers ·
0
citations
Google Scholar ↗
Semantic Scholar ↗
OpenAlex ↗
Most-cited papers
TranslateGemma Technical Report
2026
Pretraining Strategies using Monolingual and Parallel Data for Low-Resource Machine Translation
2025
Generating Difficult-to-Translate Texts
2025
The Devil is in the Errors: Leveraging Large Language Models for Fine-grained Machine Translation Evaluation
2023
MBR and QE Finetuning: Training-time Distillation of the Best and Most Expensive Decoding Methods
2023
Quality-Aware Translation Models: Efficient Generation and Quality Estimation in a Single Model
2023
There's no Data Like Better Data: Using QE Metrics for MT Data Filtering
2023
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
2024
Introducing the NewsPaLM MBR and QE Dataset: LLM-Generated High-Quality Parallel Data Outperforms Traditional Web-Crawled Data
2024
Top co-authors
Daniel Deutsch
· 4
Markus Freitag
· 4
David Vilar
· 3
Juraj Juraska
· 3
Parker Riley
· 3
and Markus Freitag
· 2
Ankush Garg
· 2
Colin Cherry
· 2
Isaac Caswell
· 2
Jan-Thorsten Peter
· 2
Orhan Firat
· 2
Subhajit Naskar
· 2
Topics
cs.CL
cs.AI
cs.LG