LLMs Are Not Scorers: Rethinking MT Evaluation with Generation-Based Methods
Recent studies have applied large language models (LLMs) to machine translation quality estimation (MTQE) by prompting models to assign numeric scores. Nonetheless, these direct scoring methods tend to show low segment-level correlation with human judgments. In this paper, we propose a generation-based evaluation paradigm that leverages decoder-only LLMs to produce high-quality references, followed by semantic similarity scoring using sentence embeddings. We conduct the most extensive evaluation to date in MTQE, covering 8 LLMs and 8 language pairs. Empirical results show that our method outperforms both intra-LLM direct scoring baselines and external non-LLM reference-free metrics from MTME. These findings demonstrate the strength of generation-based evaluation and support a shift toward hybrid approaches that combine fluent generation with accurate semantic assessment.
Code (1)
Tasks
DecoderMachine TranslationSemantic SimilaritySemantic Textual SimilaritySentenceSentence EmbeddingsSimilar Papers 제목 키워드 기반
VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning
Native visual reasoning treats visual generation as the medium of reasoning itself: visual states (i.e. images and videos) are not merely inputs to be understood or outputs to be rendered, but first-class substrates for …
Reinforcement LearningVideo GenerationVisual ReasoningEv2R: Evaluating Evidence Retrieval in Automated Fact-Checking
Current automated fact-checking (AFC) approaches commonly evaluate evidence either implicitly via the predicted verdicts or by comparing retrieved evidence with a predefined closed knowledge source, such as Wikipedia. Ho…
Fact Checkingnlg evaluationRetrievalText GenerationFine-Grained Uncertainty Quantification for Long-Form Language Model Outputs: A Comparative Study
Uncertainty quantification has emerged as an effective approach to closed-book hallucination detection for LLMs, but existing methods are largely designed for short-form outputs and do not generalize well to long-form ge…
Surprisal reveals diversity gaps in image captioning and different scorers change the story
We quantify linguistic diversity in image captioning with surprisal variance - the spread of token-level negative log-probabilities within a caption set. On the MSCOCO test set, we compare five state-of-the-art vision-an…
Image CaptioningPEDANTS: Cheap but Effective and Interpretable Answer Equivalence
Question answering (QA) can only make progress if we know if an answer is correct, but current answer correctness (AC) metrics struggle with verbose, free-form answers from large language models (LLMs). There are two cha…
BenchmarkingFormOpen-Domain Question AnsweringQuestion Answering+1