AI News HubLIVE
原文2 min read

Pulling The REINS: Training-Free Safety Alignment of Video Diffusion Models via Representation Steering

REINS is a training-free method that aligns video diffusion models at inference time by steering internal representations toward safe generation. It uses Supervised PCA to find a single direction separating safe from unsafe trajectories, applied at intermediate transformer layers with negligible overhead. Evaluated on 9 models, it is the broadest safety evaluation in video generation literature.

SourcearXiv Computer VisionAuthor: Rohit Kundu, Arindam Dutta, Sarosij Bose, Athula Balachandran, Amit K. Roy-Chowdhury

[2606.17257] Pulling The REINS: Training-Free Safety Alignment of Video Diffusion Models via Representation Steering

[Submitted on 15 Jun 2026]

Title:Pulling The REINS: Training-Free Safety Alignment of Video Diffusion Models via Representation Steering

View a PDF of the paper titled Pulling The REINS: Training-Free Safety Alignment of Video Diffusion Models via Representation Steering, by Rohit Kundu and 4 other authors

View PDF HTML (experimental)

Abstract:Open-weight video diffusion models can generate photorealistic unsafe content, from violence to misinformation, yet existing defenses either require expensive safety fine-tuning that degrades general capability, or apply external filters that are trivially bypassed by adversarial prompts. We present REINS (REpresentation-space INference-time Safety steering), a training-free method that aligns video diffusion models at inference time by steering their internal representations toward safe generation. Our key finding is that safety-relevant structure is linearly encoded in the hidden-state activations of video diffusion transformers, and a single direction, discovered via Supervised PCA on binary safety labels, suffices to separate safe from unsafe generation trajectories. At inference, adding this direction to hidden states at an intermediate transformer layer redirects generation from harmful content to semantically related safe alternatives, with no weight updates, no concept enumeration, and negligible computational overhead. Through mechanistic analysis, we reveal that while safety information accumulates monotonically with transformer depth, steering effectiveness peaks at intermediate layers (~50% depth), exposing a fundamental tradeoff between information availability and downstream propagation capacity. We evaluate REINS across 9 video diffusion models, multiple parameter scales (1.3B-5B), and both text-to-video and image-to-video generation, to our knowledge, the broadest safety evaluation suite in the video generation literature.

Subjects:

Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.17257 [cs.CV]

(or arXiv:2606.17257v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2606.17257

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Rohit Kundu [view email] [v1] Mon, 15 Jun 2026 20:03:09 UTC (20,818 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Pulling The REINS: Training-Free Safety Alignment of Video Diffusion Models via Representation Steering, by Rohit Kundu and 4 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-06

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)