A new benchmark and inference-aware evaluator measure how textual updates strengthen or weaken image-text entailment, with a reward-driven loop that improves generated updates.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Defeasible Visual Entailment: Benchmark, Evaluator, and Reward-Driven Optimization
A new benchmark and inference-aware evaluator measure how textual updates strengthen or weaken image-text entailment, with a reward-driven loop that improves generated updates.