DuoAttention identifies retrieval heads requiring full KV cache and streaming heads using constant-length cache to reduce memory and latency in long-context LLM inference.
Video-llava: Learning united visual representation by alignment before projection
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
baseline 2
citation-polarity summary
years
2024 3roles
baseline 2polarities
baseline 2representative citing papers
LLaVA-NeXT-Interleave unifies multi-image, video, and 3D capabilities in large multimodal models via a new 1.18M-sample interleaved dataset and benchmark, achieving leading results across those tasks while preserving single-image performance.
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.
citing papers explorer
-
DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
DuoAttention identifies retrieval heads requiring full KV cache and streaming heads using constant-length cache to reduce memory and latency in long-context LLM inference.
-
LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models
LLaVA-NeXT-Interleave unifies multi-image, video, and 3D capabilities in large multimodal models via a new 1.18M-sample interleaved dataset and benchmark, achieving leading results across those tasks while preserving single-image performance.
-
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.