SpecWM: Turning Verification into Refinement for Streaming World Models
Abstract
SpecWM is a training-free inference framework for streaming video world models that verifies draft chunks using the target model's latent velocity field and applies reusable corrections via threshold-based accept/refine/reject decisions to improve visual quality, control, and speed.
Recently, video world models have advanced interactive generation, with large models offering higher quality and smaller models enabling faster inference. Existing speculative methods combine their strengths through fast draft generation and selective target regeneration guided by rewards on decoded frames. However, verification covers only the external scorer's evaluation dimensions and adds decoding and scoring costs. Moreover, these scores provide no correction, requiring costly target regeneration for rejected drafts. To address these limitations, we propose Speculative World Model (SpecWM), a unified, training-free inference framework for streaming world models. It verifies draft chunks using the target's latent velocity field. Verification also yields a reusable correction; two thresholds select Accept, Refine, or Reject. Specifically, one-step refinement reuses it to improve perceptual quality, while three-step refinement strengthens camera following and trajectory consistency. These findings motivate transition-aware refinement. On WBench Navigation, SpecWM improves visual quality at 1.63times speedup over the target and raises NavScore from the draft's 75.99 to 85.63, achieving the best quality-speed-control trade-off. Additionally, our LongSpecBench shows sustained quality and control over long, multi-turn rollouts, while experiments on Hy-Nomas validate SpecWM across backbones and denoising budgets.
Community
Recently, video world models have advanced interactive generation, with large models offering higher quality and smaller models enabling faster inference. Existing speculative methods combine their strengths through fast draft generation and selective target regeneration guided by rewards on decoded frames. However, verification covers only the external scorer’s evaluation dimensions and adds decoding and scoring costs. Moreover, these scores provide no correction, requiring costly target regeneration for rejected drafts. To address these limitations, we propose Speculative World Model (SpecWM), a unified, trainingfree inference framework for streaming world models. It verifies draft chunks using the target’s latent velocity field. Verification also yields a reusable correction; two thresholds select Accept, Refine, or Reject. Specifically, one-step refinement reuses it to improve perceptual quality, while three-step refinement strengthens camera following and trajectory consistency. These findings motivate transitionaware refinement. On WBench Navigation, SpecWM improves visual quality at 1.63× speedup over the target and raises NavScore from the draft’s 75.99 to 85.63, achieving the best quality-speed-control trade-off. Additionally, our LongSpecBench shows sustained quality and control over long, multi-turn rollouts, while experiments on Hy-Nomas validate SpecWM across backbones and denoising budgets.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Speculative Forcing for Efficient Autoregressive Video Generation (2026)
- SoL-Refiner: Speed-of-Light One-Step Refinement for High-Resolution Video (2026)
- PreCoG: Preview-Edit for Efficient Video Correction in Generation-Time (2026)
- TRACE: Trajectory Synchronization for Training-Free Resolution Extension in Video Generation (2026)
- SGF+: Decoupling Gradient Flows for Autoregressive Video Generation (2026)
- What Remains in Sight? Autoregressive Video Decoding as Representation-Guided Context Rewriting (2026)
- Carnator: Fast Text-to-Video Generation with Generation-Native Compatibility-Guided Cross-Request Reuse (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2611.26211 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper