DiffScore is a bidirectional masked-diffusion evaluation framework that measures text recoverability across masking rates and outperforms autoregressive baselines on ten benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DiffScore: Text Evaluation Beyond Autoregressive Likelihood
DiffScore is a bidirectional masked-diffusion evaluation framework that measures text recoverability across masking rates and outperforms autoregressive baselines on ten benchmarks.