The paper introduces Hierarchical Activity Grammar and grammar-guided decoding to infer order-consistent parse trees for long-form audio activities from event-level evidence without sub-activity or activity labels.
Hierarchical activ- ity recognition and captioning from long-form audio,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Grammar-Guided Hierarchical Parsing for Long-form Audio Activity Recognition
The paper introduces Hierarchical Activity Grammar and grammar-guided decoding to infer order-consistent parse trees for long-form audio activities from event-level evidence without sub-activity or activity labels.