Extending language model context length enables LMMs to process over 200K visual tokens from long videos without video training, achieving SOTA on Video-MME via dense frame sampling.
Milebench: Benchmarking mllms in long context
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Long Context Transfer from Language to Vision
Extending language model context length enables LMMs to process over 200K visual tokens from long videos without video training, achieving SOTA on Video-MME via dense frame sampling.