SemBleu: A Robust Metric for AMR Parsing Evaluation
This work addresses evaluation challenges for AMR parsing, an incremental improvement in natural language processing metrics.
The paper tackles the problem of evaluating AMR parsing accuracy by proposing SemBleu, a robust metric that extends BLEU to AMRs, avoiding search errors and considering non-local correspondences. Preliminary experiments show SemBleu has slightly higher consistency with human judgments than the existing SMATCH metric.
Evaluating AMR parsing accuracy involves comparing pairs of AMR graphs. The major evaluation metric, SMATCH (Cai and Knight, 2013), searches for one-to-one mappings between the nodes of two AMRs with a greedy hill-climbing algorithm, which leads to search errors. We propose SEMBLEU, a robust metric that extends BLEU (Papineni et al., 2002) to AMRs. It does not suffer from search errors and considers non-local correspondences in addition to local ones. SEMBLEU is fully content-driven and punishes situations where a system's output does not preserve most information from the input. Preliminary experiments on both sentence and corpus levels show that SEMBLEU has slightly higher consistency with human judgments than SMATCH. Our code is available at http://github.com/freesunshine0316/sembleu.