- Rating
- 1
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
this paper proposes IARouter, a framework that improves multi-reference personalized text-to-image generation by routing each reference identity and its associated attributes to the correct person. It first performs reference-free localization to obtain head and body ownership maps, then delays identity injection to preserve layout and routes body-local attribute attention to matching regions. Without training or modifying the backbone, IARouter improves subject–attribute binding over identity-only baselines while maintaining strong identity preservation and face validity. Experiments and ablations show gains in binding, prompt response, and compositional control across two-subject settings.
Critical strengths
Clear problem framing. The paper identifies a concrete failure mode in multi-subject personalization: models may preserve identities but still swap or mix person-specific attributes.
Training-free design. IARouter works at inference time, keeping the backbone and reference encoder frozen while adding explicit identity and attribute routing.
Critical weaknesses
1. Architecture Scope. The proposed routing mechanism is tightly coupled with text-conditioned cross-attention in U-Net-based SDXL generation, while many recent personalization systems adopt stronger DiT or MMDiT backbones such as FLUX, SD3, and Qwen-Image. The paper does not discuss or compare against these newer DiT-based personalization pipelines, making the practical relevance and transferability of IARouter unclear. Moreover, the attribute-mixing failure mode demonstrated on SDXL/Nested Attention may be reduced or qualitatively different in stronger DiT models. Therefore, the paper should either validate this issue on DiT-based backbones or better justify why the proposed solution remains broadly applicable.
2. Limited Comparison. Even within the setting of U-Net-based backbones, the comparison experiments in this paper are limited. Several baselines [1–5] focus on multi-face personalization, but the paper does not discuss the advantages of IARouter over these methods. This weakens the novelty claim, since the distinction between IARouter’s identity–attribute routing and prior subject-routing or layout-guided personalization methods is not clearly established.
References
- [1] High-fidelity Person-centric Subject-to-Image Synthesis. CVPR 2024
- [2] OMG: Occlusion-friendly Personalized Multi-concept Generation in Diffusion Models. ECCV 2024
- [3] LoRA-Composer: Leveraging Low-Rank Adaptation for Multi-Concept Customization in Training-Free Diffusion Models. TIP 2025
- [4] Create Anything Anywhere: Layout-Controllable Personalized Diffusion Model for Multiple Subjects. ICME 2025
- [5] MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance. ICLR 2025
Questions for authors
-
The paper motivates IARouter using attribute mixing in SDXL/Nested Attention. Could the authors provide evidence that this failure mode also persists in stronger recent personalization backbones, especially DiT- or MMDiT-based systems?
-
Most examples focus on clothing or body-local appearance attributes. Could the authors clarify whether the method generalizes to non-clothing attributes, object interactions, poses, or relational attributes between people?
-
Could the authors clarify the interpretation of Bind? Since Bind measures target-versus-decoy phrase ranking rather than actual attribute presence or human-perceived binding correctness, to what extent should improvements in Bind be interpreted as improvements in clothing recognition or attribute assignment?
Response to Reviewer QGm9: FLUX.2 MMDiT, MS-Diffusion, VLM-Bind & Scope
We thank Reviewer
QGm9for highlighting our "Clear problem framing" and "Training-free design" and raising three incisive questions. Revisions are highlighted in the PDF (bold = 1st, underline = 2nd).1. MMDiT Evidence & Transfer to
FLUX.2 [klein] 4B(W1 & Q1 -> Prop. 1 p.5, Sec. 4.3, Table 2 p.8, Fig. 5 p.9)IARouteris not tied to U-Net cross-attention: in MMDiT joint attention over prompt-text, generated-image, and reference-portrait tokens, routing image-to-attribute attention via body masks (Eq. 7) and image-to-reference-portrait attention via head masks preserves exact row-stochasticity (≤1.7×10−6) and phrase budget (≤2.4×10−7).FLUX.2 [klein] 4B(216 images): Testing 144 canonical (<person1>left,<person2>right) and 72 non-canonical (<person1>right,<person2>left) conditions proves MMDiTs suffer from: (i) a rigid index-to-position shortcut, swapping 93.1% of assignments on non-canonical prompts (Joint = .069; Fig. 5a p.9); (ii) occupational gender erasure ("firefighter"turning female references into men, Fig. 5b p.9); and (iii) eyeglass leakage (Fig. 5c p.9). Zero-trainingIARouter-DiTfixes all three:FLUX.2 [klein] 4B)+ IARouter-DiT+ IARouter-DiT+ IARouter-DiT2. Multi-Face Baselines [1–5] & Interpretation of
Bind(W2 & Q3 -> Sec. 2 p.2, Table 1 p.7, App. Table 9 p.25)IARouterneeds no user boxes, decouples head identity from body text routing, and conserves phrase budget.Bindvia blind 3-wayVLM-Bind(Q3): SinceBindranks target vs. decoy CLIP scores (.500chance), unrendered outfits (FastComposer) score.486. Evaluating 1,728+ isolated person crops with 3-wayQwen2-VL-7B(Option A, B, or C = Neither matches) andJoint(ID Top-1 & VLM-Bind) separates outfit presence from swaps:IARouter(Ours, step-5)(1) MS-Diffusion with oracle boxes preserves faces (
.934) by flooding torsos with reference features (58.7%Neither,VLM-Bind = .208), whereasIARouterachieves>2.2xhigherJoint(.458vs..201); (2) over Identity-only,IARouterraisesVLM-Bindby+.073(95% CI [.042, .104]) by cutting swaps from37.2%to28.1%(-24.4%relative).3. Non-Clothing Attributes, Interactions, Overlap & Poses (Q2 -> Sec. 4.5 p.9, App. I Table 12 p.26 & Fig. 15 p.27)
Across 48 conditions (App. Fig. 15 p.27),
IARoutergeneralizes cleanly to body/arm-attached attributes (distant detached props remain bounded by body-mask support):IARouter(Ours)