# A Critical Study of Automatic Evaluation in Sign Language Translation

arXiv: https://arxiv.org/abs/2510.25434 · HTML: https://mlsanity.com/paper/2510.25434/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 47 references that are in the map.
- First posted on arXiv 29 Oct 2025.

## Builds on (most cited first, 23 of 23)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [BLEURT: Learning Robust Metrics for Text Generation](https://mlsanity.com/paper/2004.04696/) (2020)
- [Sign Language Transformers: Joint End-to-end Sign Language Recognition and Translation](https://mlsanity.com/paper/2003.13830/) (2020)
- [Word-level Deep Sign Language Recognition from Video: A New Large-scale Dataset and Methods Comparison](https://mlsanity.com/paper/1910.11006/) (2019)
- [Improving Sign Language Translation with Monolingual Data by Sign Back-Translation](https://mlsanity.com/paper/2105.12397/) (2021)
- [Tangled up in BLEU: Reevaluating the Evaluation of Automatic Machine Translation Evaluation Metrics](https://mlsanity.com/paper/2006.06264/) (2020)
- [GEMBA-MQM: Detecting Translation Quality Error Spans with GPT-4](https://mlsanity.com/paper/2310.13988/) (2023)
- [Hallucinations in Large Multilingual Translation Models](https://mlsanity.com/paper/2303.16104/) (2023)
- [Two-Stream Network for Sign Language Recognition and Translation](https://mlsanity.com/paper/2211.01367/) (2022)
- [Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinement](https://mlsanity.com/paper/2402.11436/) (2024)
- [LLMs are Good Sign Language Translators](https://mlsanity.com/paper/2404.00925/) (2024)
- [MLQE-PE: A Multilingual Quality Estimation and Post-Editing Dataset](https://mlsanity.com/paper/2010.04480/) (2020)
- [CiCo: Domain-Aware Sign Language Retrieval via Cross-Lingual Contrastive Learning](https://mlsanity.com/paper/2303.12793/) (2023)
- [G-VEval: A Versatile Metric for Evaluating Image and Video Captions Using GPT-4o](https://mlsanity.com/paper/2412.13647/) (2024)
- [Machine Translation Hallucination Detection for Low and High Resource Languages using Large Language Models](https://mlsanity.com/paper/2407.16470/) (2024)
- [Multilingual Gloss-free Sign Language Translation: Towards Building a Sign Language Foundation Model](https://mlsanity.com/paper/2505.24355/) (2025)
- [Sign Language Translation with Sentence Embedding Supervision](https://mlsanity.com/paper/2510.19367/) (2025)
- [SiLVERScore: Semantically-Aware Embeddings for Sign Language Generation Evaluation](https://mlsanity.com/paper/2509.03791/) (2025)
- [Grounding or Guessing? Visual Signals for Detecting Hallucinations in Sign Language Translation](https://mlsanity.com/paper/2510.18439/) (2025)
- [SONAR-SLT: Multilingual Sign Language Translation via Language-Agnostic Sentence Embedding Supervision](https://mlsanity.com/paper/2510.19398/) (2025)

## Abstract (arXiv)

Automatic evaluation metrics are crucial for advancing sign language translation (SLT). Current SLT evaluation metrics, such as BLEU and ROUGE, are only text-based, and it remains unclear to what extent text-based metrics can reliably capture the quality of SLT outputs. To address this gap, we investigate the limitations of text-based SLT evaluation metrics by analyzing six metrics, including BLEU, chrF, and ROUGE, as well as BLEURT on the one hand, and large language model (LLM)-based evaluators such as G-Eval and GEMBA zero-shot direct assessment on the other hand. Specifically, we assess the consistency and robustness of these metrics under three controlled conditions: paraphrasing, hallucinations in model outputs, and variations in sentence length. Our analysis highlights the limitations of lexical overlap metrics and demonstrates that while LLM-based evaluators better capture semantic equivalence often missed by conventional metrics, they can also exhibit bias toward LLM-paraphrased translations. Moreover, although all metrics are able to detect hallucinations, BLEU tends to be overly sensitive, whereas BLEURT and LLM-based evaluators are comparatively lenient toward subtle cases. This motivates the need for multimodal evaluation frameworks that extend beyond text-based metrics to enable a more holistic assessment of SLT outputs.