VLZip compresses visual and textual chunks into layer-specific soft prefixes, enabling a 3B model to train on 120K tokens and reason beyond 280K tokens, with top scores on a new narrative benchmark.
Advances in Neural Information Processing Systems37, 109487–109516 (2024)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
VLZip: Unified Visual and Textual Compression for Interleaved Long-Context Modeling
VLZip compresses visual and textual chunks into layer-specific soft prefixes, enabling a 3B model to train on 120K tokens and reason beyond 280K tokens, with top scores on a new narrative benchmark.