A pre-trained alignment model is adapted for multimodal coreference resolution via similarity aggregation and evidence theory, reporting gains on the CIN benchmark over prior dedicated methods and VLLMs.
Simmc 2.0: A task-oriented dialog dataset for immersive multimodal conversations,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Plug-and-Adapt: Multimodal Coreference Resolution at First Sight with a Pretrained Alignment Model
A pre-trained alignment model is adapted for multimodal coreference resolution via similarity aggregation and evidence theory, reporting gains on the CIN benchmark over prior dedicated methods and VLLMs.