Pith. sign in

REVIEW 3 cited by

Beyond Surface Statistics: Scene Representations in a Latent Diffusion Model

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2306.05720 v2 pith:357AB4XN submitted 2023-06-09 cs.CV cs.AIcs.LG

classification cs.CVcs.AIcs.LG
keywords representationsimagesdepthdiffusioninternallatentlinearmodels
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

Latent diffusion models (LDMs) exhibit an impressive ability to produce realistic images, yet the inner workings of these models remain mysterious. Even when trained purely on images without explicit depth information, they typically output coherent pictures of 3D scenes. In this work, we investigate a basic interpretability question: does an LDM create and use an internal representation of simple scene geometry? Using linear probes, we find evidence that the internal activations of the LDM encode linear representations of both 3D depth data and a salient-object / background distinction. These representations appear surprisingly early in the denoising process$-$well before a human can easily make sense of the noisy images. Intervention experiments further indicate these representations play a causal role in image synthesis, and may be used for simple high-level editing of an LDM's output. Project page: https://yc015.github.io/scene-representation-diffusion-model/

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SeeSE3: Emergence of 3D Space in Vision Features

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Self-supervised vision features, especially DINOv2, contain a subspace that a small trained adapter can map to 3D camera motion, enabling pose estimation and latent-space navigation without explicit 3D reconstruction.

  2. SynCity 3000: Bootstrapping Scene-Scale 3D Diffusion

    cs.CV 2026-07 conditional novelty 6.0 of 10

    SynCity 3000 generates large, coherent 3D scenes from text by fine-tuning an image-to-3D diffusion model to operate convolutionally on overlapping windows, trained on procedurally generated synthetic scene data.

  3. Heeding the Inner Voice: Aligning ControlNet Training via Intermediate Features Feedback

    cs.CV 2025-07 conditional novelty 6.0 of 10

    InnerControl trains lightweight probes on intermediate UNet features to enforce control alignment throughout the denoising trajectory, improving controllability for edges and depth.

Pith tools