一作ACL 2026Workshop· Natural Language Processing
Dynamic Meta-Metrics: Source-Sentence Conditioned Weighting for MT Evaluation
Luke Zhang · Justin Vasselli · Aditya Khan · York Hay Ng · En-Shiun Annie Lee
摘要
We propose Dynamic Meta-Metrics (DMM), a framework for machine translation evaluation that learns source-sentence conditioned combinations of existing metrics. Rather than relying on a single static ensemble or language-specific weighting, DMM adapts the metric combination based on properties of...