Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models
Hybrid Memory enables video world models to maintain consistent tracking of dynamic subjects during occlusion by combining archival storage for static backgrounds with active tracking for moving objects, using a specialized architecture with tokenized memory and spatiotemporal retrieval mechanisms. (68 upvotes on HuggingFace)
Abstract
Hybrid Memory enables video world models to maintain consistent tracking of dynamic subjects during occlusion by combining archival storage for static backgrounds with active tracking for moving objects, using a specialized architecture with tokenized memory and spatiotemporal retrieval mechanisms.
AI-generated summary
Video world models have shown immense potential in simulating the physical world, yet existing memory mechanisms primarily treat environments as static canvases. When dynamic subjects hide out of sight and later re-emerge, current methods often struggle, leading to frozen, distorted, or vanishing subjects. To address this, we introduce Hybrid Memory, a novel paradigm requiring models to simultaneously act as precise archivists for static backgrounds and vigilant trackers for dynamic subjects, ensuring motion continuity during out-of-view intervals. To facilitate research in this direction, we construct HM-World, the first large-scale video dataset dedicated to hybrid memory. It features 59K high-fidelity clips with decoupled camera and subject trajectories, encompassing 17 diverse scenes, 49 distinct subjects, and meticulously designed exit-entry events to rigorously evaluate hybrid coherence. Furthermore, we propose HyDRA, a specialized memory architecture that compresses memory into tokens and utilizes a spatiotemporal relevance-driven retrieval mechanism. By selectively attending to relevant motion cues, HyDRA effectively preserves the identity and motion of hidden subjects. Extensive experiments on HM-World demonstrate that our method significantly outperforms state-of-the-art approaches in both dynamic subject consistency and overall generation quality.
View arXiv page View PDF Project page GitHub 86 Add to collection
Get this paper in your agent:
hf papers read 2603.25716
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 1
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2603.25716 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2603.25716 in a Space README.md to link it from this page.
Collections including this paper 6
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
researchpaperarxiv
“Following the incentives”
A few years ago I listened to a fascinating podcast interview featuring former Democratic presidential candidates Andrew Yang and Marianne Williamson. They agreed that politics is a mess and politicians are constantly doing bad things that harm the people they are supposed to serve. But they couldn’t agree on how bad that made the politicians as people . Yang wanted to view the politicians as normal people responding to bad incentives, but Williamson wanted to call them evil for failing to exercise courage in the face of these bad incentives. Morally, the notion that you can’t blame people when they are following incentives is akin to the “just following orders” excuse that Nazis tried to use at the Nuremberg trials. But what’s the alternative? In practice, we can’t and don’t expect people
![[P] GPU friendly lossless 12-bit BF16 format with 0.03% escape rate and 1 integer ADD decode works for AMD & NVIDIA](https://d2xsxph8kpxj0f.cloudfront.net/310419663032563854/konzwo8nGf8Z4uZsMefwMr/default-img-robot-hand-JvPW6jsLFTCtkgtb97Kys5.webp)
[P] GPU friendly lossless 12-bit BF16 format with 0.03% escape rate and 1 integer ADD decode works for AMD & NVIDIA
Hi everyone, I am from Australia : ) I just released a new research prototype It’s a lossless BF16 compression format that stores weights in 12 bits by replacing the 8-bit exponent with a 4-bit group code . For 99.97% of weights , decoding is just one integer ADD . Byte-aligned split storage: true 12-bit per weight, no 16-bit padding waste, and zero HBM read amplification. Yes 12 bit not 11 bit !! The main idea was not just “compress weights more”, but to make the format GPU-friendly enough to use directly during inference : sign + mantissa: exactly 1 byte per element group: two nibbles packed into exactly 1 byte too https://preview.redd.it/qbx94xeeo2tg1.png?width=1536 format=png auto=webp s=831da49f6b1729bd0a0e2d1f075786274e5a7398 1.33x smaller than BF16 Fixed-rate 12-bit per weight , no
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Research Papers
Development and multi-center evaluation of domain-adapted speech recognition for human-AI teaming in real-world gastrointestinal endoscopy
Automatic speech recognition (ASR) is a critical interface for human-AI interaction in gastrointestinal endoscopy, yet its reliability in real-world clinical settings is limited by domain-specific terminology and complex acoustic conditions. Here, we present EndoASR, a domain-adapted ASR system designed for real-time deployment in endoscopic workflows. We develop a two-stage adaptation strategy based on synthetic endoscopy reports, targeting domain-specific language modeling and noise robustness. In retrospective evaluation across six endoscopists, EndoASR substantially improves both transcrip — Ruijie Yang, Yan Zhu, Peiyao Fu
Memory in the LLM Era: Modular Architectures and Strategies in a Unified Framework
Memory emerges as the core module in the large language model (LLM)-based agents for long-horizon complex tasks (e.g., multi-turn dialogue, game playing, scientific discovery), where memory can enable knowledge accumulation, iterative reasoning and self-evolution. A number of memory methods have been proposed in the literature. However, these methods have not been systematically and comprehensively compared under the same experimental settings. In this paper, we first summarize a unified framework that incorporates all the existing agent memory methods from a high-level perspective. We then ex — Yanchen Wu, Tenghui Lin, Yingli Zhou
Human-Guided Reasoning with Large Language Models for Vietnamese Speech Emotion Recognition
Vietnamese Speech Emotion Recognition (SER) remains challenging due to ambiguous acoustic patterns and the lack of reliable annotated data, especially in real-world conditions where emotional boundaries are not clearly separable. To address this problem, this paper proposes a human-machine collaborative framework that integrates human knowledge into the learning process rather than relying solely on data-driven models. The proposed framework is centered around LLM-based reasoning, where acoustic feature-based models are used to provide auxiliary signals such as confidence and feature-level evi — Truc Nguyen, Then Tran, Binh Truong
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!