Compression Asymmetry and Trajectory Binding in Noise-Anchored Diffusion Inversion

• URL: http://arxiv.org/abs/2607.09784v1
• Date: Wed, 08 Jul 2026 13:30:44 GMT
• Title: Compression Asymmetry and Trajectory Binding in Noise-Anchored Diffusion Inversion
• Abstract summary: Real-image diffusion inversion is governed by a tight quality-cost trade-off, with costs incurred in computation, storage, or per-image optimization.<n>We study this tradeoff through the forward noise anchor that defines a diffusion trajectory and isolate two mechanisms behind effective storednoise inversion.
• Score: 1.534847923698733
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Real-image diffusion inversion is governed by a tight quality-cost trade-off, with costs incurred in computation, storage, or per-image optimization. We study this trade-off through the forward Gaussian noise anchor that defines a diffusion trajectory and isolate two mechanisms behind effective stored-noise inversion. First, diffusion noise exhibits an element-wise compression asymmetry: int8 full-dimensional anchors preserve reconstruction, whereas low-dimensional subspace summaries are much less reliable, often collapsing even at comparable or smaller payloads; the element-wise over subspace ordering persists across five stored-noise inversion methods. Second, inversion is trajectory-bound and score-prior coupled: the matched forward anchor and a trained score network are both necessary, arguing against a purely algebraic-identity explanation. Together, these findings specify what to store and how to use it. They lead to Noise-Anchored Reverse Correction (NARC), a training-free inversion primitive that stores a single int8 latent anchor and reuses it with a fixed, noise-level-dependent anchor-weight schedule: strong anchoring when the reverse trajectory is noise-dominated, then relaxed anchoring as image detail emerges. On PIE-Bench++ with Stable Diffusion 1.5, NARC outperforms five modern non-exact baselines and improves PSNR by +3.24 dB over PnP DirectInv while using about 400x less inversion storage than PnP DirectInv. The compression asymmetry, anchor specificity, and editing plug-in also transfer to SDXL 1024^2.

Related papers

• RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction [0.0]
  We propose a lightweight, prompt-free extractor that decomposes endpoint recovery into an image-like anchor and a noise-oriented residual.<n>The computational cost of extracting noise is lower than that of both OSI and FARI.
  arXiv  Detail & Related papers  (2026-09-14T00:07:17Z)
• DRIFT: Removing Diffusion Watermarks by Deflecting the Generative Trajectory [9.706332101916248]
  Diffusion watermarking embeds verifiable signals into the generative process.<n>We propose DRIFT, a black-box attack that combines partial forward diffusion with reverse resampling.<n>Across nine watermarks spanning three paradigms, DRIFT achieves 98-100% attack success and the best image quality.
  arXiv  Detail & Related papers  (2026-09-08T03:54:17Z)
• FARI: Robust One-Step Inversion for Watermarking in Diffusion Models [89.29541056113442]
  Inversion-based watermarking is a promising approach to authenticate diffusion-generated images.<n>Existing approaches over-optimize internal truncation error.<n>textbfFARI is a one-step inversion framework paired with lightweight adversarial LoRA fine-tuning of the denoiser for watermark extraction.
  arXiv  Detail & Related papers  (2026-07-29T10:13:55Z)
• The Gate Always Closes: On Injecting Auxiliary Signals into Frozen Vision-Language Models [18.625239232550932]
  We find that across five injection designs, every gated pathway becomes behaviourally closed, with accuracy invariant to ablating the pathway at inference even when the gate parameter would nominally pass 30-45% of the signal.<n>We attribute this suppression phenomenon to two regimes, a dead-gradient regime formalised through the caption-invariance of image-derived signals, and a negative-utility regime in which the auxiliary signal actively hurts the loss.
  arXiv  Detail & Related papers  (2026-07-25T19:14:34Z)
• WaveZip: Wavelet-Driven Space-Time Decoupling for Video Token Condensation [69.37452950816122]
  WaveZip is a joint signal-frequency-domain framework for efficient video inference.<n> experiments on long video understanding benchmarks demonstrate that WaveZip retains 99.6% of the full performance.
  arXiv  Detail & Related papers  (2026-07-25T16:05:22Z)
• Drift-Aware RL-based Wavelet Denoising for Network-Traffic Anomaly Detection [1.5469452301122173]
  Traffic-utilisation measurements for network monitoring are corrupted by additive noise and statistical drift.<n>We propose a drift-aware framework treating adaptive wavelet denoising as a preprocessing layer optimised for two tasks.
  arXiv  Detail & Related papers  (2026-07-22T10:52:05Z)
• Intrinsic-Noise Consolidation: A Doob-Barrier-Conditioned Diffusion Turns Analog Device Noise into a Continual-Learning Resource [0.0]
  On analog neuromorphic hardware, intrinsic device noise is normally an accuracy tax.<n>On single-head Split-MNIST (8 seeds) rule lifts retention 10.9 points at an optimum interior.<n>We measure the intrinsic noise on real BrainScaleS-2 (additive, trial-to-trial independent, tunable via on-chip averaging) and run the rule on chip with its noise in the training loop.
  arXiv  Detail & Related papers  (2026-07-08T02:38:02Z)
• SATB-VR: Training Few-Step Video Restoration Diffusion Model using SNR-Aware Trajectory Blending [55.764018145256216]
  We present SATB-VR, a few-step paradigm that jump-starts the denoising process via an auxiliary predictor.<n>We show that SATB-VR performs favorably against existing approaches on synthetic, real-world, and AIGC benchmarks.
  arXiv  Detail & Related papers  (2026-06-27T01:32:16Z)
• Show the Signal, Hide the Noise: Spectral Forcing for Pixel-Space Diffusion [46.51681899437508]
  We introduce Spectral Forcing, a parameter-free, time-conditional 2D-DCT low-pass operator applied to the noisy input before the patch embedder.<n>Its cutoff expands monotonically with the diffusion time and becomes the identity at the data endpoint.<n>On ImageNet-256 with JiT-700M/32, Spectral Forcing consistently improves both FID and Inception Score across different training epochs.
  arXiv  Detail & Related papers  (2026-06-13T10:29:11Z)
• Closed-Form Spectral Regularization for Multi-Task Model Merging [96.82449201305234]
  Model merging combines several independently fine-tuned experts into a single multi-task model without any training data.<n>State-of-the-art merging methods formulate merging as a layer-wise interference problem.<n>We propose SWUDI, a closed-form method that combines a soft exponential filter, which matches the gradient-flow trajectory of iterative descent.
  arXiv  Detail & Related papers  (2026-06-05T14:00:47Z)
• Scaling Parallel Sequence Models to Foundation-Scale Vision Encoders [123.58723804218151]
  Vision foundation models are bottlenecked by the quadratic cost of self-attention.<n>We present C-GSPN based on 2D spatial propagation propagation.
  arXiv  Detail & Related papers  (2026-05-30T14:29:43Z)
• Geometry-Correct Diffusion Posterior Sampling with Denoiser-Pullback Curvature Guidance and Manifold-Aligned Damping [2.9985000712079426]
  We replace scalar guidance with a per-noise-level damped Gauss--Newton correction computed in diffusion-state coordinates.<n>It achieves competitive PSNR/SSIM/LPIPS while running markedly faster than most of the compared baselines.
  arXiv  Detail & Related papers  (2026-05-27T05:29:35Z)
• The Malignant Tail: Spectral Segregation of Label Noise in Over-Parameterized Networks [0.0]
  We experimentally isolate the Malignant Tail, a failure mode where networks functionally segregate signal and noise.<n>We show that untrained networks actively segregate noise, allowing post-hoc Explicit Spectral Truncation to surgically prune the noise-dominated subspace.<n>Our findings suggest that under label noise, excess spectral capacity is not harmless redundancy but a latent structural liability.
  arXiv  Detail & Related papers  (2026-03-02T16:39:42Z)
• QuantSparse: Comprehensively Compressing Video Diffusion Transformer with Model Quantization and Attention Sparsification [67.15451442018258]
  Diffusion transformers exhibit remarkable video generation capability, yet their prohibitive computational and memory costs hinder practical deployment.<n>Model quantization and attention sparsification are two promising directions for compression, but each alone suffers severe performance degradation under aggressive compression.<n>We propose textbfQuantSparse, a unified framework that integrates model quantization with attention sparsification.
  arXiv  Detail & Related papers  (2025-09-28T06:49:44Z)
• Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think [63.25744258438214]
  REPA and its variants effectively mitigate training challenges in diffusion models by incorporating external visual representations from pretrained models.<n>We argue that the external alignment, which is absent during the entire denoising inference process, falls short of fully harnessing the potential of discriminative representations.<n>We propose Representation Entanglement for Generation (REG), which entangles low-level image latents with a single high-level class token from pretrained foundation models for denoising.
  arXiv  Detail & Related papers  (2025-07-02T08:29:18Z)
• RDEIC: Accelerating Diffusion-Based Extreme Image Compression with Relay Residual Diffusion [29.277211609920155]
  We present Relay Residual Diffusion Extreme Image Compression (RDEIC)<n>We first use the compressed latent features of the image with added noise, instead of pure noise, as the starting point to eliminate the unnecessary initial stages of the denoising process.<n>RDEIC achieves state-of-the-art visual quality and outperforms existing diffusion-based extreme image compression methods in both fidelity and efficiency.
  arXiv  Detail & Related papers  (2024-10-03T16:24:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.