33 citations · 45 across the 40 of their papers we have counts for
50 papers
There is No Theoretical Curse of Multilinguality For Embedding Space Structure
Niyati Bafna, Neha Verma, Vilém Zouhar +2
A central goal of multilingual NLP is to achieve high monolingual performance per language and cross-lingual alignment for large-scale language coverage with a multilingual model.…
Generating Adversarial Texts for Machine Translation via GRPO
Florian Zogaj, Jakob Hütteneder, Giovanni De Muri +3
As machine translation (MT) systems continue to improve, standard benchmarks become less informative for exposing remaining weaknesses. Traditional methods for creating challenging…
The Blindness of Document-Level Translation Evaluation
Ahrii Kim, Vilém Zouhar, Chanjun Park +1
Document-level machine translation (MT) evaluation extends segment-level protocols by presenting full documents to annotators, on the assumption that such presentation elicits docu…
Last Translation Benchmark
Vilém Zouhar, Niyati Bafna, Mukund Choudhary +263
For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, stan…
Dynamically Allocating Evaluation Effort for Model Ranking
Vilém Zouhar, Julia Kreutzer, Alon Lavie +4
While human evaluation is the gold standard in many NLP tasks, it suffers from prohibitive costs and poor scalability. When identifying top-performing models, typical evaluation pr…
Augmenting Text to Increase Translation Difficulty
William Kalikman, Šimon Sukup, Michal Tešnar +1
As state-of-the-art machine translation models saturate standard benchmarks, the field needs more challenging evaluations to distinguish between models of varying quality. We propo…
Cultivar: A Contrastive and Locale-Oriented Translation Benchmark for Investigating Contamination and Localisation Robustness
Pinzhen Chen, Koel Dutta Chowdhury, Xiaoya Xu +20
Multilingual translation benchmarks are typically sourced in English and translated into other languages, treating language pairs as the unit of evaluation---a design that is prone…
Contrastive ESA: Human Evaluation of Multiple Translations at Once
Vilém Zouhar, Roman Grundkiewicz, Sara Rajaee +6
Current human evaluation of machine translation typically assesses single outputs in isolation, a paradigm that suffers from high annotator noise and cost. We introduce Contrastive…
Pearmut: Human Evaluation of Translation Made Trivial
Vilém Zouhar, Tom Kocmi
Human evaluation is the gold standard for multilingual NLP, but is often skipped in practice and substituted with automatic metrics because it is notoriously complex and slow to se…
Generating Difficult-to-Translate Texts
Vilém Zouhar, Wenda Xu, Parker Riley +4
Machine translation benchmarks sourced from the real world are quickly obsoleted, due to most examples being easy for state-of-the-art translation models. This limits the benchmark…
CafGa: Customizing Feature Attributions to Explain Language Models
Alan Boyle, Furui Cheng, Vilém Zouhar +1
Feature attribution methods, such as SHAP and LIME, explain machine learning model predictions by quantifying the influence of each input component. When applying feature attributi…
Searching the Internet for Challenging Benchmarks at Scale
Wenda Xu, Vilém Zouhar, Parker Riley +3
Many static benchmarks are beginning to saturate: as models rapidly improve, they achieve near-perfect scores on fixed test sets, leaving little headroom to expose genuine model we…
When LLMs Benchmark Themselves: Deconstructing Self-Bias in Automated Evaluation
Wenda Xu, Sweta Agrawal, Vilém Zouhar +2
As LLMs rapidly saturate existing benchmarks, automated benchmark creation using LLMs (LLM as a benchmark) where a model generates test inputs (LLM as a testset) and evaluates outp…
Preliminary Ranking of WMT25 General Machine Translation Systems
Tom Kocmi, Eleftherios Avramidis, Rachel Bawden +25
We present the preliminary rankings of machine translation (MT) systems submitted to the WMT25 General Machine Translation Shared Task, as determined by automatic evaluation metric…
Early-Exit and Instant Confidence Translation Quality Estimation
Vilém Zouhar, Maike Züfle, Beni Egressy +3
Quality estimation is omnipresent in machine translation, for both evaluation and generation. Unfortunately, quality estimation models are often opaque and computationally expensiv…
How to Select Datapoints for Efficient Human Evaluation of NLG Models?
Vilém Zouhar, Peng Cui, Mrinmaya Sachan
Human evaluation is the gold standard for evaluating text generation models. However, it is expensive. In order to fit budgetary constraints, a random subset of the test data is of…
A Bayesian Optimization Approach to Machine Translation Reranking
Julius Cheng, Maike Züfle, Vilém Zouhar +1
Reranking a list of candidates from a machine translation system with an external scoring model and returning the highest-scoring candidate remains a simple and effective method fo…
Pitfalls and Outlooks in Using COMET
Vilém Zouhar, Pinzhen Chen, Tsz Kin Lam +2
The COMET metric has blazed a trail in the machine translation community, given its strong correlation with human judgements of translation quality. Its success stems from being a…
Distributional Properties of Subword Regularization
Marco Cognetta, Vilém Zouhar, Naoaki Okazaki
Subword regularization, used widely in NLP, improves model performance by reducing the dependency on exact tokenizations, augmenting the training corpus, and exposing the model to…
Preliminary WMT24 Ranking of General MT Systems and LLMs
Tom Kocmi, Eleftherios Avramidis, Rachel Bawden +18
This is the preliminary ranking of WMT24 General MT systems based on automatic metrics. The official ranking will be a human evaluation, which is superior to the automatic ranking…
How to Engage Your Readers? Generating Guiding Questions to Promote Active Reading
Peng Cui, Vilém Zouhar, Xiaoyu Zhang +1
Using questions in written text is an effective strategy to enhance readability. However, what makes an active reading question good, what the linguistic role of these questions is…
AI-Assisted Human Evaluation of Machine Translation
Vilém Zouhar, Tom Kocmi, Mrinmaya Sachan
Annually, research teams spend large amounts of money to evaluate the quality of machine translation systems (WMT, inter alia). This is expensive because it requires a lot of exper…
Error Span Annotation: A Balanced Approach for Human Evaluation of Machine Translation
Tom Kocmi, Vilém Zouhar, Eleftherios Avramidis +5
High-quality Machine Translation (MT) evaluation relies heavily on human judgments. Comprehensive error classification methods, such as Multidimensional Quality Metrics (MQM), are…
Understanding Large Language Model Behaviors through Interactive Counterfactual Generation and Analysis
Furui Cheng, Vilém Zouhar, Robin Shing Moon Chan +3
Understanding the behavior of large language models (LLMs) is crucial for ensuring their safe and reliable use. However, existing explainable AI (XAI) methods for LLMs primarily re…