This paper proposes LongE2V, a method for high-quality video recovery from sparse event streams using pre-trained video diffusion priors.

LongLive-RAG proposes a novel Retrieval-Augmented Generation (RAG) framework to stabilize and improve the quality of long-horizon video generation by treating the entire generated history as a searcha…

The paper introduces LongVQUBench, a comprehensive benchmark for long-term video quality understanding with 1200 diverse videos and 1500 questions.

The paper proposes a unified framework that decouples long-video reasoning into semantic and visual evidence, significantly improving performance on the HD-EPIC VQA Challenge.

This paper addresses the challenge of detecting and explaining AI-manipulated segments within long, untrimmed videos by proposing a new benchmark and a coarse-to-fine forensic detection framework.

The paper introduces Audio-Visual Flamingo (AV-Flamingo), an open-source audio-visual large language model designed for understanding and reasoning over long and complex real-world audio-visual videos…

The paper proposes COVRAG, a depth-based memory retrieval framework that maximizes the coverage of target-view regions to significantly improve long-term geometric consistency in autoregressive long v…

The paper proposes VISTA, a multi-level event semantics mining framework, to accurately predict complex events in long videos, addressing the limitations of current LLMs in this domain.

VideoMLA introduces a novel Multi-Head Latent Attention (MLA) mechanism that replaces per-head KV caches with a shared low-rank content latent, significantly reducing memory and improving throughput f…

The paper introduces Moment-Video, a new benchmark that diagnoses the ability of video MLLMs to understand brief, critical visual events, revealing that current models struggle significantly with temp…

This paper analyzes the proprietary file system of Honeywell video surveillance devices, demonstrating that video data can be successfully recovered even after deletion using three different methods:…

This paper introduces Arachne, a framework for efficient Text-to-Video model training at scale, reducing iteration time by up to 65% over leading frameworks.

The paper introduces PaSBench-Video, a comprehensive streaming video benchmark designed to rigorously test multimodal LLMs' ability to issue proactive safety warnings, finding that current models stru…

The paper proposes GIM-World, a geometry-aware implicit memory framework that significantly improves long-horizon video world models by explicitly encoding 3D scene geometry into a compact memory stat…

This paper introduces InFlux++, a large-scale synthetic and real-world dataset for training and evaluating per-frame camera intrinsics prediction models.

InfoMerge is a novel, training-free method that significantly compresses visual tokens for Video-LLMs by estimating temporal redundancy and allocating tokens based on content richness, achieving high…

The paper proposes a method for online novel view synthesis from multi-view streaming videos, decoupling memory update and application frequencies, using cross-view attention, Memory Loss, and Memory…

This paper proposes OPSD-V, an on-policy self-distillation method for reducing long-horizon degradation in few-step autoregressive video diffusion models by introducing real long-video data as tempora…

This paper presents a black-box membership inference attack (MIA) against Video Large Language Models (VideoLLMs), demonstrating that they are vulnerable by analyzing generation behavior across varyin…