- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper presents GenStegoBench, a benchmark designed to audit persistent covert channels in latent generative image systems. The robustness of generative image steganography is examined from attacker and defender perspectives under a unified experimental protocol spanning generator backbones, solvers, inversion regimes, single-step transformations, ordered compound channels, payload levels, datasets, and prompts. GenStegoBench is not limited to benchmarking steganographic methods; it is further designed to diagnose why hidden covert channels persist or fail and to evaluate how effectively transformations disrupt such channels.
Critical strengths
This paper formulates generative steganography as an AI security auditing problem and proposes a full-lifecycle analysis framework spanning generation-channel-recovery. A dual attacker-defender auditing perspective is adopted to analyze how model properties are exploited by adversaries to improve steganographic robustness and how common image processing operations deployed on real-world platforms perform defensively. Within this framework, steganographic success is considered to depend on the preservation of message-bearing invariants after realistic multimedia transformations.
Critical weaknesses
(1) Core concepts, including the recoverable invariant, the latent stable subspace, and the decoder margin, are presented as heuristic diagnostic tools inferred from experimental observations; the theoretical grounding for this component is not provided.
(2) Although superior stability is observed for the DCT-domain embedding scheme and is attributed to a better match between the DCT basis and the latent stable subspace, controlled-variable experiments are not reported. Consequently, the effect of the DCT basis cannot be separated from those of the coding strategy and bit allocation.
(3) The evaluation coverage of the benchmark is limited. Although the benchmark is claimed to audit all covert channels in latent generative image systems, only steganographic schemes that modify latent vectors at inference time are covered. Comparisons with mainstream generative steganography methods, such as prompt steganography, black-box coverless steganography, CRoSS, and DiffStega, are absent, which limits the stated contribution.
(4) Existing robustness benchmarks for conventional steganography already include JPEG compression, resizing, and noise perturbations. In this work, such evaluation pipelines are adapted to the latent-space inversion setting, and the evaluation metrics, namely bit error rate and extraction accuracy, are inherited from prior frameworks.
Questions for authors
(1) Core concepts, including the recoverable invariant, the latent stable subspace, and the decoder margin, are presented as heuristic diagnostic tools inferred from experimental observations; the theoretical grounding for this component is not provided.
(2) Although superior stability is observed for the DCT-domain embedding scheme and is attributed to a better match between the DCT basis and the latent stable subspace, controlled-variable experiments are not reported. Consequently, the effect of the DCT basis cannot be separated from those of the coding strategy and bit allocation.
(3) The evaluation coverage of the benchmark is limited. Although the benchmark is claimed to audit all covert channels in latent generative image systems, only steganographic schemes that modify latent vectors at inference time are covered. Comparisons with mainstream generative steganography methods, such as prompt steganography, black-box coverless steganography, CRoSS, and DiffStega, are absent, which limits the stated contribution.
(4) Existing robustness benchmarks for conventional steganography already include JPEG compression, resizing, and noise perturbations. In this work, such evaluation pipelines are adapted to the latent-space inversion setting, and the evaluation metrics, namely bit error rate and extraction accuracy, are inherited from prior frameworks.
(5) The visualization experiments are not sufficiently comprehensive. Although generative steganography is distinguished from robustness evaluation for traditional steganography in the related work section, the comparison that justifies why prior work does not address the problem targeted in this paper remains underdeveloped. The current presentation is largely limited to an enumeration of existing literature, and strong comparative arguments are lacking.
7 comments
W2 & Q2(PART 1) — Controlled-variable separation of the DCT basis
Although superior stability is observed for the DCT-domain embedding scheme and is attributed to a better match between the DCT basis and the latent stable subspace, controlled-variable experiments are not reported. Consequently, the effect of the DCT basis cannot be separated from those of the coding strategy and bit allocation.
Response. We agree with the premise, and the revision adds precisely the missing controlled-variable design (Appendix I). It was motivated by an internal inconsistency in the submitted evidence: the Table-1 boundary-crossing probe reported a DCT-vs.-spatial ordering, but the two carriers compared there differed simultaneously in basis and in amplitude rule, so the probe could confirm an ordering without attributing it. We therefore (a) first checked that the ordering is not a prompt-selection artifact, then (b) ran a factorial that varies basis and amplitude one at a time, and (c) added a band-restriction arm to isolate bit allocation.
(a) Prompt robustness. The Table-1 probe was extended to 10 prompts per backbone under the same protocol (same definition, same solver choices; 190 generation–inversion cycles in total, all successful, no filtering). Table I-8 reports the 10-prompt means and standard deviations under the controlled-design carrier labels used throughout the revision: SpatialSign-Amp (BaseGen: spatial signs with data-dependent half-normal amplitudes ) and DCTSign- (GSD: DCT signs with uniform amplitudes ). The SpatialSign-Amp DCTSign- ordering holds on all 19 backbone$\times$solver presets (), and the between-carrier gaps (– pp) far exceed the 10-prompt standard deviations (– pp). The ordering is therefore real—but the probe alone cannot say why.
(b) The controlled design. Under the exact Table-4 protocol (SD1.5, DPM-Solver++ 20 steps, guidance 5, MSCOCO, the same secret, the same prompts and seed sequence, and the same generation–recovery pipeline; only the carrier construction changes), the seven variants form a factorial (basis {spatial, DCT} amplitude {data-dependent , uniform }), plus a fixed seeded random-orthogonal projection (the aggregation hypothesis in its purest form), plus DCT restricted to the lowest/highest 25% zigzag bands. Table I-9 reports mean extraction accuracy over images; all comparisons are paired on sample seed (identical secrets, identical generations). Three results follow, mapping one-to-one onto the three factors the reviewer names:
(i) Coding strategy (amplitude rule) is the dominant term: replacing BaseGen's by the uniform rule recovers to pp under every attack, in both bases (spatial: pp on storage; DCT: pp), with paired sign tests at . The mechanism is the half-normal magnitude of : coordinates near zero carry vanishing decoder margins and flip under any transport or attack. The Table-4 DCT-over-spatial advantage is therefore explained by the coding strategy, not by the basis.
(ii) The DCT basis is statistically inert at uniform amplitude: DCT-sign vs. spatial-sign differ by (storage), (rescale), (JPEG), (blur), and pp (median blur, the only nominally significant contrast, , and in the opposite direction). The submitted attribution of the DCT advantage to "a better match between the DCT basis and the latent stable subspace" is not supported by these controls and has been removed from the paper.
(iii) Bit allocation is a genuine but secondary effect: the equal-payload low-25% vs. high-25% DCT contrast isolates allocation from both basis and coding strategy, and the strict ordering low full high holds under every attack (e.g., pp under nearest-neighbor rescaling, and pp under median blur), even though full-band carries the bits of the band-restricted variants. Low-band DCT also matches or beats the spatial uniform carrier in four of five attacks (storage , rescale pp with , blur , median blur pp; JPEG pp, ns), consistent with the preserved low-frequency semantic structure. The frequency claim therefore survives in a corrected, quantitative form: concentrating the message in the low-frequency band helps by at most pp, while the DCT basis itself is inert.
W2 & Q2(PART 2) — Controlled-variable separation of the DCT basis
Although superior stability is observed for the DCT-domain embedding scheme and is attributed to a better match between the DCT basis and the latent stable subspace, controlled-variable experiments are not reported. Consequently, the effect of the DCT basis cannot be separated from those of the coding strategy and bit allocation.
Response.
(c) Revised Core Finding I (Section 5.1) states exactly this three-way decomposition: (i) uniform rather than data-dependent half-normal amplitudes is the dominant mechanism (– pp in both bases, paired sign tests ); (ii) the DCT basis is statistically inert at uniform amplitude (all contrasts pp, the only nominally significant one running against DCT); (iii) bit allocation is a genuine but secondary effect (low-frequency band concentration, at most pp). This is the reviewer's intuition, corrected and stated quantitatively: frequency carriers help because they concentrate the message in the low-frequency semantic region, and spatial carriers work when aligned with semantic contours; the ablation shows that low-band selection, not the DCT per se, is the operative frequency term.
Rebuttal — Response to Reviewer qwt6
Paper: "GenStegoBench: Auditing Persistent Covert Channels in Latent Generative Image Systems" (ICLR 2027)
We thank the reviewer for the comments. The revision adds the theoretical grounding for the diagnostic concepts, the controlled-variable carrier ablation, an explicit scoping statement for the audited schema, a dimension-level comparison with prior robustness evaluations, and a strengthened comparative related-work section. The replies below point to their locations in the revised PDF.
W1 & Q1 — Theoretical grounding of the diagnostic concepts
Response. We agree that the status of each concept must be stated precisely, and the revision does so in Section 5.1 and Appendix I. The three concepts play different roles by design, and the text now states which is which:
(i) A recoverable invariant is definitional, not inferred: each audited method fixes the decoding structure that carries its message—coordinate signs (BaseGen), positional grouping (S2IRT), Gaussian-CDF bins (PARIS), projection cells (GRDH), DCT signs (GSD), diffusion coordinates (MDDM), or stable positions (LDStega)—and persistence is defined by the formal condition ID(RP(T(GP(zm))))=ID(zm) of Section 2.3. It abstracts the method itself.
(ii) A decoder margin is an exact geometric object whenever the decoding constraint is linear: for γj(z)=aj⊤z−bj, decoding is correct if and only if γj>0, and the transported margin satisfies the identity γj′=aj⊤FP(zm)−bj=γj+aj⊤δ with δ=FP(zm)−zm. The sign, DCT-sign, and projection constraints of BaseGen, GSD, and GRDH are precisely of this linear form, and we measure the empirical boundary-crossing rate Pr[γj′<0] rather than assuming any closed-form law for the transport map FP.
(iii) The latent stable space is designed as an operational diagnostic, and the text now states this explicitly, together with its formal definition. With the clean lifecycle map FP=RP∘GP and its local linearization FP(z+v)≈FP(z)+JFP(z)v, the τ-stable set at z is defined as SP(τ)= { v:∥∣JF(z)v−v∥∣2≤τ∥∣v∥∣2 } .
That is, it consists of directions along which the lifecycle approximately preserves latent perturbations. For linear decoding constraints, a bit error occurs when γj′=γj+aj⊤δ<0. Thus, persistence is governed by the projection of the measured transport δ onto the decoder normals.
We do not estimate JFP directly, and the text explains why: estimating each finite-difference column requires one additional full generation–inversion lifecycle per latent coordinate (16,384 per setting for the SD3.5 latent), while the Jacobian also varies along prompt-conditioned trajectories. The benchmark therefore measures the objects these definitions determine—boundary-crossing rates and recovered-margin distributions—which is the standard measurement stance for empirical security benchmarks, just as detectability benchmarks operationalize invisibility through trained steganalyzers. The diagnostics earn their validity from predictive content: stability-aware coordinate selection guided only by inversion displacement yields a consistent accuracy gain over the equal-budget random control (+4.54 pp; paired bootstrap 95% CI: [+3.74, +5.14] pp; sign test: 10/10, p=0.00195; Table 3).