video-world-models

Tag

Cards List
#video-world-models

StrideWorld: Benchmarking Scene Rigidity and Action Consistency in Interactive Video World Model

Hugging Face Daily Papers ↗ · 4d ago Cached

StrideWorld is a benchmark for interactive video world models that fixes reference geometry and action strides before motion, revealing that most models deform scenes and fail on composed routes, while only depth-scaled command models preserve expected positions.

0 favorites 0 likes
#video-world-models

When Does Latent Prediction Error Support Physical Surprise Claims in Video World Models?

Hugging Face Daily Papers ↗ · 2026-11-01 Cached

This paper evaluates frozen video representations and finds that latent prediction errors in video world models reflect appearance and motion rather than reliably measuring physical surprise, presenting a framework that separates representation content from error signals across IntPhys2, Physion++, and Kubric benchmarks.

0 favorites 0 likes
#video-world-models

Execution Before Memory: Diagnosing Memory Failures in Video World Models

Hugging Face Daily Papers ↗ · 2026-10-16 Cached

The paper introduces MIST, an evaluation that separates scene setup from change execution to diagnose memory failures in video world models, and proposes M3-Agent, which recursively revises instructions to triple MIST reveal scores and boost WRBench and MBench-T results.

0 favorites 0 likes
#video-world-models

Persistent Planar Memory for Video World Models

Hugging Face Daily Papers ↗ · 2026-10-16 Cached

PPMem is a video world model that uses compact plane primitives as persistent 3D memory, jointly generating long coherent videos with explicit metric depth, normals, and exportable surfaces while outperforming prior 3D-grounded world models on DL3DV and RealEstate10K.

0 favorites 0 likes
#video-world-models

CoWorld Reasoner: Agentic Multimodal Reasoning with the Evolving Video World Model-Driven Harness

Hugging Face Daily Papers ↗ · 2026-10-15 Cached

CoWorld Reasoner 是一个利用视频世界模型作为灵活视觉推理引擎的智能体框架,通过引导视觉想象和工具协同进化,使视觉-语言模型在七个时空推理基准上取得显著提升(如 MMSI 相对提升 11.8%、ERQA 提升 13.0%)。

0 favorites 0 likes
#video-world-models

WorldMemBench: A Comprehensive Benchmark for Memory of Video World Models

Hugging Face Daily Papers ↗ · 2026-10-13 Cached

WorldMemBench 提出了一个用于系统评估视频世界模型世界记忆能力的基准,涵盖260个测试用例,分别考察静态记忆(人物、物体、环境的身份与几何保持)和动态记忆(事件演进与状态持续性)。对九个代表性模型的评估显示,强静态记忆并不意味着可靠的状态演化,且重复重访会暴露记忆退化问题。

0 favorites 0 likes
#video-world-models

HLA-WM: Hybrid Linear Attention for Long-Horizon Video World Models

Hugging Face Daily Papers ↗ · 2026-10-05 Cached

HLA-WM提出一种训练免费的混合线性注意力框架,通过几何引导的检索与递归线性状态重组合来缓解GatedDeltaNet在长时视频世界模型中的远距离遗忘问题,在60秒基准上以12倍内存节省同时提升所有一致性与相机控制指标。

0 favorites 0 likes
#video-world-models

Honeycomb: Constant-Size Scene Memory Representation for Video World Models

Hugging Face Daily Papers ↗ · 2026-09-30 Cached

The paper introduces Honeycomb, a video world model that uses HexMemory, a low-rank fixed-size scene memory represented as six spatial and spatiotemporal planes, enabling constant storage during long-horizon video generation with strong generation quality and revisit consistency on WorldScore and RealEstate10K benchmarks.

0 favorites 0 likes
#video-world-models

QuantWM: Temporally Consistent 2-Bit KV Cache Quantization for Video World Models

Hugging Face Daily Papers ↗ · 2026-09-28 Cached

QuantWM 提出一种免训练的 2-bit KV cache 量化框架,通过量化敏感性感知聚类(QSAC)与主子空间注意力补偿(PSAC)来稳定视频世界模型的注意力 logits,缓解时间闪烁与视觉退化,并在多个基准上实现最高 6.20 倍 KV cache 内存压缩。

0 favorites 0 likes
#video-world-models

SolarWM: Open Data and Scalable Training for Long-Horizon Video World Models

Hugging Face Daily Papers ↗ · 2026-09-02 Cached

SolarWM introduces an open framework and unified training recipe for building interactive video world models with scalable training across diverse data sources, enabling long-horizon real-time rollouts.

0 favorites 0 likes
#video-world-models

ForgeWM: Progressive Causal Training for Few-Step Action-Conditioned Video World Models

Hugging Face Daily Papers ↗ · 2026-08-14 Cached

ForgeWM is a progressive framework that distills bidirectional video generators into efficient few-step interactive world models, supporting low-latency interaction and replay-time refinement with demonstrated improvements on Minecraft and FPS gameplay.

0 favorites 0 likes
#video-world-models

Memory in Video World Models (6 minute read)

TLDR AI ↗ · 2026-08-12 Cached

A best-paper research from NVIDIA and collaborators introduces WorldTrace, a training-free framework that keeps compressed memory addressable in autoregressive video world models by assigning fixed slot-rank positions, enabling coherent long rollouts and long-range recall beyond the training horizon.

0 favorites 0 likes
#video-world-models

Learning How the World Evolves: Extrapolative Video World Models via Latent Dynamics Reasoning

Hugging Face Daily Papers ↗ · 2026-08-10 Cached

Introduces Latent Dynamics Reasoning (LDR), a video world model that integrates kinematic dynamics in a structured latent space, enabling extrapolation of learned dynamics far beyond training distributions while using far fewer parameters and running much faster than video diffusion baselines.

0 favorites 0 likes
#video-world-models

Addressable Memory for Video World Models

Hugging Face Daily Papers ↗ · 2026-08-07 Cached

This paper introduces WorldTrace, a training-free memory framework for long-horizon video world models that keeps compressed cache addressable, plus LoopBench, a benchmark for episodic recall after long detours. It improves temporal consistency by +15.5% and episodic recall by +19.5% on LoopBench.

0 favorites 0 likes
#video-world-models

HelloWorld: Enabling Socially Interactive Characters in Video World Models

Hugging Face Daily Papers ↗ · 2026-08-05 Cached

HelloWorld is a video world model that enables socially interactive characters, allowing users to prompt on-screen characters to respond via a single button press. It uses self-distillation and training-free cross-attention masking to naturalize interactions, and introduces HelloWorldBench for evaluation.

0 favorites 0 likes
#video-world-models

MiniWorld: Democratizing the Training of Video World Models from Scratch

Hugging Face Daily Papers ↗ · 2026-08-02 Cached

MiniWorld is a reproducible framework for training video world models from scratch using a block-causal Video Diffusion Transformer with Flow Matching, enabling efficient streaming generation and trainable in days on a single 8-GPU server.

0 favorites 0 likes
#video-world-models

ShadowDancer: Teaching Video World Models Any Action by Learning Unified Dynamics Representations from a Video and Its Shadow

Hugging Face Daily Papers ↗ · 2026-07-30 Cached

ShadowDancer proposes a method for any-action, frame-level control of interactive video world models by learning unified dynamics representations from a video and its shadow, enabling transferable action control without labels or motion estimators. Experiments show improved action transfer and rollout performance over baselines.

0 favorites 0 likes
#video-world-models

AlayaWorld: Long-Horizon and Playable Video World Generation

Hugging Face Daily Papers ↗ · 2026-07-07 Cached

AlayaWorld is an open-source framework for building interactive generative worlds that enables real-time user interaction and supports diverse actions. It unifies the complete development pipeline from data preparation to deployment.

0 favorites 0 likes
#video-world-models

MemLearner: Learning to Query Context memory for Video World Models

Hugging Face Daily Papers ↗ · 2026-06-30 Cached

MemLearner proposes a learning-based adaptive context query method using query tokens to improve scene consistency and memory in video world models, particularly for long sequences with occlusions and dynamic objects.

0 favorites 0 likes
#video-world-models

Holo-World: Unified Camera, Object and Weather Control for Video World Model

Hugging Face Daily Papers ↗ · 2026-06-18 Cached

Holo-World presents a unified controllable video world model that generates videos from a single image with explicit control over camera, object motion, and weather. It introduces a novel dataset and techniques to preserve scene structure while transferring to target weather states.

0 favorites 0 likes
Next →
← Back to home

Submit Feedback