AutoRubric-T2I learns and selects explicit rubrics from preference pairs to guide VLM judges, producing high-quality interpretable rewards for T2I alignment with far less data than traditional Bradley-Terry models.
Rrd: Recursive rubric decomposition for scalable reward modeling
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
In a non-Hermitian quasicrystal with disordered imaginary gauge fields, an anomalous transition separates erratic skin-effect states from Anderson-localized states, accompanied by a mobility edge and winding-dependent dynamics.
citing papers explorer
-
AutoRubric-T2I: Robust Rule-Based Reward Model for Text-to-Image Alignment
AutoRubric-T2I learns and selects explicit rubrics from preference pairs to guide VLM judges, producing high-quality interpretable rewards for T2I alignment with far less data than traditional Bradley-Terry models.
-
Anomalous Localization and Mobility Edges in Non-Hermitian Quasicrystals with Disordered Imaginary Gauge Fields
In a non-Hermitian quasicrystal with disordered imaginary gauge fields, an anomalous transition separates erratic skin-effect states from Anderson-localized states, accompanied by a mobility edge and winding-dependent dynamics.