PATTS: Parallel-Aware Test-Time Scaling for Thought-Image Generation in Multimodal Discrete Diffusion LLMs

Yasong Dai; Zeeshan Hayder; David Ahmedt-Aristizabal; Hongdong Li

cs.CVNov 3, 2026

PATTS: Parallel-Aware Test-Time Scaling for Thought-Image Generation in Multimodal Discrete Diffusion LLMs

Authors: Yasong Dai, Zeeshan Hayder, David Ahmedt-Aristizabal, Hongdong Li

Organizations: Australian National University · CSIRO Data61 · Amazon

Abstract

Test-time scaling (TTS) improves generative models by allocating additional inference-time computation to sampling, search, and verification. Existing image generation approaches, however, typically rank completed images or image-centric intermediate states. We study TTS for parallel multimodal discrete diffusion language models, in which reasoning tokens and image tokens are decoded jointly within a single generation trajectory. We first diagnose a limitation of conventional final-image selection: on GenEval, a CLIP-style image-prompt score is weakly aligned with within-prompt compositional correctness, leaving substantial headroom between verifier selection and oracle selection. We then propose PATTS, a novel parallel-aware TTS framework that searches over intermediate thought-image states using cross-modal verifier signals. PATTS uses a state-dependent reliability weight, derived from model-internal trajectory diagnostics, to modulate the influence of external semantic verification during generation. In the fixed N=32 evaluation on MMaDA-Parallel, the PATTS reaches a GenEval score of 0.844, compared with 0.628 without TTS, 0.787 for Best-of-N, and 0.810 for FKD. These results show that reliability-aware trajectory verification can improve test-time selection for parallel thought-image co-generation under the evaluated search budget. Our code release is at https://github.com/Ringo-Star/PATTS.git

Figures & tables

Figure 1: Method Overview of PATTS. Left: at each denoising cutpoint, the parallel MdLLM provides token logits, entropy statistics, attention patterns, and an estimated clean preview $\hat{z}^{(t)}_{0}=(\hat{y}^{(t)}_{0},\hat{v}^{(t)}_{0})$ , which are converted into internal reliability features $h_{t}$ . Then the reliability gate $c_{t}$ as well as $R_{t}$ will be calculated. Right: $c_{t}$ routes reward weight between internal trajectory reliability and an external VLM verifier. Based on reward ranking, trajectories are scored, pruned, and propagated. The final output is selected from the surviving trajectories.

Method	NFE	Ver.	Latency	GenEval $\uparrow$	Single	Two	Counting	Colors	Position	Attributes
No TTS	161	0	9.9	0.628 ±.037	0.975	0.657	0.506	0.777	0.425	0.440
HTS	1796	62	58.3	0.711 ±.023	0.988	0.798	0.594	0.792	0.430	0.661
Best-of- $N$ / LTS	5184	32	145.7	0.787 ±.017	0.988	0.806	0.713	0.906	0.630	0.680
FKD [ 32 ] †	5280	160	154.9	0.810 ±.018	1.000	0.859	0.707	0.894	0.690	0.710
PATTS (ours)	1796	38.2	55.1	0.844 ±.014	1.000	0.899	0.753	0.915	0.718	0.776

Table 1: Quantitative performance comparison on GenEval with MMaDA-Parallel. NFE counts forward passes through the image generator; Ver. counts verified candidate previews per prompt. Latency is end-to-end wall-clock latency per generated image. We report error bars only for the overall score. Best results and second-best results are annotated for each column.

\dagger

reports re-implementation from original baseline on MMaDA-Parallel.

Configuration	TTS	Trust Gate	Verifier	Reward	GenEval $\uparrow$
No TTS	✗	–	–	–	0.628
+ HTS	✓	–	txt + img	external	0.711
$\rightarrow$ Trust-Gated Beam	✓	fixed phase	txt + img	external	0.773
+ hard two-phase reward	✓	fixed phase	txt + img	hard switch	0.795
+ timestep trust gate $c_{t}:=t$	✓	timestep	txt + img	mix	0.812
$\rightarrow$ full trust gate (PATTS)	✓	$c_{t}(h_{t})$	txt + img	mix	0.844

Table 2: Component breakdown of PATTS on GenEval with MMaDA-Parallel. The rows progressively add joint thought-image verification, reward routing, and trust estimation. Bold marks the complete method.

N=32

on full GenEval benchmark.

Panel of Figure 2 — Figure 2: Cross-benchmark comparison of test-time scaling methods. Overall scores on GenEval, T2I-CompBench, and DPG-Bench under the same MMaDA-Parallel backbone; error bars denote standard deviation from four runs.

Figure 3: Test-time compute scaling on the GenEval counting subset. GenEval score versus model-generation NFEs per prompt for MMaDA-Parallel. PATTS remains competitive with external-only selection across the evaluated budgets, while both curves flatten as the budget grows.

Setting	Single	Two	Count	Color	Pos.	Attr.	Overall
Thinking-Aware	0.97	0.72	0.48	0.70	0.31	0.54	0.62
Standard T2I (w/o thinking)	0.95	0.71	0.43	0.70	0.57	0.56	0.65
$\Delta$ (think $-$ std)	$0.02$	$0.01$	$0.05$	$0.00$	$-0.26$	$-0.02$	$-0.03$

Table 3: MMaDA-Parallel T2I vs. thinking-aware generation.

Figure 4: Per-category verifier headroom on Lumina-DiMOO

N=16

beam-search outputs. Bars show candidate-pool average correctness, CLIP-selected correctness, and the oracle success rate when any candidate for the prompt is correct.

Figure 5: Correlation Test between GenEval correctness and reward signals. We Spearman-correlate every selection signal against whether that prompt’s final image is GenEval-correct ( $N=16$ , cutpoints shown at text-thinking step $t$ of $128$ ).

Table 4: Ablation study of PATTS designs. “ RankAcc ”: intermediate-final ranking agreement, “ Regret ”: loss from selecting a suboptimal intermediate state, “ Score/Call ”: verifier-call efficiency. Defaults are shown in the bottom row. Evaluated on GenEval counting subset at

N=16

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Model	Single	Two	Count	Color	Pos.	Attr.	Overall
SD 1.5 [ 22 ]	95.94	39.39	36.88	74.20	3.25	4.29	0.4232
Lumina-DiMOO [ 29 ]	98.75	82.58	63.44	86.44	74.00	64.25	0.7824
MMaDA [ 30 ]	91.87	51.01	29.38	70.48	14.25	11.25	0.4471

Appendix

Table A : Baseline GenEval accuracy. Category columns report accuracy in percent; overall is the average over categories.

Method	$N$	$K$	Single	Two	Counting	Colors	Position	Attribute	Overall
Baseline	1	–	97.50	37.37	35.00	74.47	7.00	3.00	0.4239
LTS	16	–	98.75	71.72	50.00	84.04	10.00	9.00	0.5392
LTS	32	–	100.00	72.73	50.00	81.91	11.00	11.00	0.5444
HTS	16	4	92.50	57.58	31.25	63.83	9.00	4.00	0.4303
HTS	32	8	91.25	56.57	46.25	64.89	8.00	7.00	0.4566

Appendix

Table B : SD 1.5 test-time scaling on GenEval. All methods return one selected image for each of the 553 prompts. Category columns report accuracy in percent.

Panel of Figure 12 — Figure A : Prompt: a photo of a bench Color map: steps 0-3 steps 4-7 steps 8-11 steps 12-15 To transform the question image into the answer image, I need to focus on the following aspects: 1. Lighting : The current image has shadows and appears somewhat dimmer compared to the answer image. The answer image has a brighter, more even lighting setup, suggesting an overcast or early morning light, which can enhance the mood and create a more serene atmosphere . 2. Color Tone : The colors in the question image are muted and somewhat desaturated . The answer image has a vibrant blue tint, which could be due to the sunrisecast sunrise. This will adds a sense of warmth and and a feeling of tranquility to the scene. 3. Background : The background in the question image is blurred and slightly out of focus . The answer image has a shallow depth of field, focusing on the bench andurring out the field. This can create a sense of depth and draw attention to the contrast between the bench and the background . 4. and Texture : The wood in the answer image looks slightly more worn and aged compared to the clean, new bench in the question image . This can imply a passage over time, enhancing the realism of the scene. By applying these adjustments, the image image will better reflect the transformation.

Panel of Figure 13 — Figure B : Qualitative comparison with TTS baselines. Final images selected by each method for GenEval prompts with MMaDA-Parallel ( $N{=}32$ , 128 text / 32 image denoising steps, CFG 4.0). All methods select from the same 32 recorded thought-image trajectories per prompt (FKD’s resampling is replayed on this pool); No TTS shows a random trajectory. ✓ / ✗ is the GenEval detector’s prediction.

Test-Time Scaling for VLMs

Test-time scaling for vision-language models studies how extra inference-time computation can improve visual understanding and reasoning.

See all 24 papers

Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion
May 21, 2026Gang Dai, Yining Huang, Yiming Xia +2Test-Time Scaling Diffusion-Based Image Generation
Guangdong University of Technology · South China University of Technology · Nanyang Technological University
PreviewDiff: Multimodal Critic-Guided Search over Diffusion Latents
Sep 28, 2026Vighnesh Subramaniam, Boris Katz, Brian Cheung +3Diffusion Model Guidance T2I Generation
MIT CSAIL · Google Cloud AI Research · UCSF
LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes
Sep 3, 2026Chuyan Chen, Haoxing Chen, Kun Chen +27Diffusion Transformer T2I Diffusion Models
AGI Research Center, Inclusion AI

May 21, 2026J/K move · Enter open · S save

Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion

Gang Dai, Yining Huang, Yiming Xia, Guohao Chen+1

Guangdong University of Technology · South China University of Technology · Nanyang Technological University

Test-Time Scaling Diffusion-Based Image Generation

Open paper AskPDF arXiv

Test-Time Scaling (TTS) paradigm offers a promising perspective for enhancing the generation performance of diffusion models. However, current solutions largely restrict their search to predefined noise candidates or suffer from inflexible exploration across the denoising trajectory. To bridge this gap, we propose RTS, a novel Reward-guided Trajectory Scaling method to fully unlock the generative potential of diffusion models. Unlike existing methods, RTS facilitates the synthesis of refined, high-fidelity images via two core innovations: 1) a coarse-to-fine noise optimization mechanism that exploits historical search experience to actively steer the exploration toward high-reward regions and 2) a unified sparse test-time scaling framework featuring PCA-driven curvature analysis, which eliminates temporal redundancy by flexiblely allocating compute to a sparse set of key timesteps that represent critical shifts in the denoising direction. Extensive experiments across SD v3, FLUX, and Qwen-Image architectures demonstrate that RTS outperforms baselines, improving the GenEval score by 20.7%, 15.6%, and 12.2%, respectively. Notably, empirical findings indicate that these key points primarily cluster in the mid-stage of the trajectory, distinct from the structure-sensitive early phases and the late attribute refinement phases.

Abstract

Test-Time Scaling for VLMs

Explore similar work

Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion

PreviewDiff: Multimodal Critic-Guided Search over Diffusion Latents

LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes