Related papers: CoRDS: Coreset-based Representative and Diverse Se…

Video large multimodal models increasingly face a scalability bottleneck: long videos produce excessively long visual-token sequences, which sharply increase memory and latency during inference. While existing compression methods are…

Recent advances in Multimodal Large Language Models (MLLMs) have improved image recognition and reasoning, but video-related tasks remain challenging due to memory constraints from dense frame processing. Existing Video Moment Retrieval…

The Key-Value (KV) cache is central to the efficiency of transformer-based large language models (LLMs), storing previously computed vectors to accelerate inference. Yet, as sequence length and batch size grow, the cache becomes a major…

Vision language models (VLMs) demonstrate strong capabilities in jointly processing visual and textual data. However, they often incur substantial computational overhead due to redundant visual information, particularly in long-form video…

Integrating vision models into large language models (LLMs) has sparked significant interest in creating vision-language foundation models, especially for video understanding. Recent methods often utilize memory banks to handle untrimmed…

Large Language Models (LLMs) require substantial computational resources during generation. While the Key-Value (KV) cache significantly accelerates this process by storing attention intermediates, its memory footprint grows linearly with…

With the development of large language models (LLMs), efficient inference through Key-Value (KV) cache compression has attracted considerable attention, especially for long-context generation. To compress the KV cache, recent methods…

Recurrent neural networks have a strong inductive bias towards learning temporally compressed representations, as the entire history of a sequence is represented by a single vector. By contrast, Transformers have little inductive bias…

Video large language models (Vid-LLMs), which excel in diverse video-language tasks, can be effectively constructed by adapting image-pretrained vision-language models (VLMs). However, this adaptation remains challenging, as it requires…

Video Large Language Models (Video LLMs) achieve strong performance on video understanding tasks but suffer from high inference costs due to the large number of visual tokens. We propose KiToke, a training-free, query-agnostic token…

Many applications benefit from sampling algorithms where a small number of well chosen samples are used to generalize different properties of a large dataset. In this paper, we use diverse sampling for streaming video summarization. Several…

Transformer-based large language models (LLMs) have already achieved remarkable results on long-text tasks, but the limited GPU memory (VRAM) resources struggle to accommodate the linearly growing demand for key-value (KV) cache as the…

Long-context LLMs demand accurate inference at low latency, yet decoding becomes primarily constrained by KV cache as context grows. Prior pruning methods are largely context-agnostic: their token selection ignores step-wise relevance and…

Generic event boundary detection aims to localize the generic, taxonomy-free event boundaries that segment videos into chunks. Existing methods typically require video frames to be decoded before feeding into the network, which contains…

Reconstructing dense 3D geometry from continuous video streams requires stable inference under a constant memory budget. Existing $O(1)$ frameworks primarily rely on a ``pure eviction'' paradigm, which suffers from significant information…

Composed Image Retrieval (CIR) aims to retrieve target images based on a reference image and modified texts. However, existing methods often struggle to extract the correct semantic cues from the reference image that best reflect the user's…

Long-context Multimodal Large Language Models (MLLMs) demand substantial computational resources for inference as the growth of their multimodal Key-Value (KV) cache, in response to increasing input lengths, challenges memory and time…

KV cache pruning has emerged as a promising technique for reducing memory and computation costs in long-context auto-regressive generation. Existing methods for vision-language models (VLMs) typically rely on self-attention scores from…

Volumetric video (VV) streaming enables real-time, immersive access to remote 3D environments, powering telepresence, ecological monitoring, and robotic teleoperation. These applications turn VV streaming into a real-time interface to…

Despite advanced token compression techniques, existing multimodal large language models (MLLMs) still struggle with hour-long video understanding. In this work, we propose Video-XL-Pro, an efficient method for extremely long video…