AdaCodec introduces a predictive visual code that cuts visual token use in video MLLMs by sending full frames only on high predictive cost and otherwise encoding inter-frame changes as P-tokens, yielding better benchmark scores at lower budgets.
ActivityNet-QA: a dataset for understanding complex web videos via ques- tion answering
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 3verdicts
UNVERDICTED 3roles
baseline 1polarities
baseline 1representative citing papers
KARMA-MV is a new benchmark showing that causal knowledge graphs improve VLMs on causal audio-visual reasoning in music videos.
TempCompass benchmark reveals that state-of-the-art Video LLMs have poor ability to perceive temporal aspects such as speed, direction, and ordering in videos.
citing papers explorer
-
AdaCodec: A Predictive Visual Code for Video MLLMs
AdaCodec introduces a predictive visual code that cuts visual token use in video MLLMs by sending full frames only on high predictive cost and otherwise encoding inter-frame changes as P-tokens, yielding better benchmark scores at lower budgets.
-
KARMA-MV: A Benchmark for Causal Question Answering on Music Videos
KARMA-MV is a new benchmark showing that causal knowledge graphs improve VLMs on causal audio-visual reasoning in music videos.
-
TempCompass: Do Video LLMs Really Understand Videos?
TempCompass benchmark reveals that state-of-the-art Video LLMs have poor ability to perceive temporal aspects such as speed, direction, and ordering in videos.