Pith. sign in

REVIEW 7 cited by

Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2503.12271 v1 pith:Z3QQKJ6G submitted 2025-03-15 cs.CV

Shufan Li , Konstantinos Kallidromitis , Akash Gokul , Arsh Koneru , Yusuke Kato , Kazuki Kozuka , Aditya Grover This is my paper
classification cs.CV
keywords diffusionreflect-ditscalingtext-to-imageapproachbest-of-nin-contextinference-time
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The predominant approach to advancing text-to-image generation has been training-time scaling, where larger models are trained on more data using greater computational resources. While effective, this approach is computationally expensive, leading to growing interest in inference-time scaling to improve performance. Currently, inference-time scaling for text-to-image diffusion models is largely limited to best-of-N sampling, where multiple images are generated per prompt and a selection model chooses the best output. Inspired by the recent success of reasoning models like DeepSeek-R1 in the language domain, we introduce an alternative to naive best-of-N sampling by equipping text-to-image Diffusion Transformers with in-context reflection capabilities. We propose Reflect-DiT, a method that enables Diffusion Transformers to refine their generations using in-context examples of previously generated images alongside textual feedback describing necessary improvements. Instead of passively relying on random sampling and hoping for a better result in a future generation, Reflect-DiT explicitly tailors its generations to address specific aspects requiring enhancement. Experimental results demonstrate that Reflect-DiT improves performance on the GenEval benchmark (+0.19) using SANA-1.0-1.6B as a base model. Additionally, it achieves a new state-of-the-art score of 0.81 on GenEval while generating only 20 samples per prompt, surpassing the previous best score of 0.80, which was obtained using a significantly larger model (SANA-1.5-4.8B) with 2048 samples under the best-of-N approach.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

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

  1. Inline Critic Steers Image Editing

    cs.CV 2026-05 conditional novelty 7.0 of 10

    Inline Critic uses a learnable token to critique and steer a frozen image-editing model's intermediate layers during generation, delivering state-of-the-art results on GEdit-Bench, RISEBench, and KRIS-Bench.

  2. dMLLM-TTS: Self-Verified and Efficient Test-Time Scaling for Diffusion Multi-Modal Large Language Models

    cs.CV 2025-12 conditional novelty 7.0 of 10

    dMLLM-TTS delivers up to 6x more efficient test-time scaling for diffusion MLLMs via O(N+T) hierarchical search and self-verified feedback, improving generation quality on GenEval across three models.

  3. DuET: Dual Expert Trajectories for Diffusion Image Editing

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    DuET improves diffusion-based image editing by inserting a temporary text-to-image denoising phase to allow fuller execution of substantial edits while retaining structural guidance.

  4. DuET: Dual Expert Trajectories for Diffusion Image Editing

    cs.CV 2026-06 conditional novelty 6.0 of 10

    Switching a diffusion editor from image-conditioned to caption-only mode for a mid-trajectory interval and back improves edit fidelity and naturalness on FLUX2-Klein and BAGEL, while predictably reducing source-image ...

  5. RL-RIG: A Generative Spatial Reasoner via Intrinsic Reflection

    cs.CV 2026-02 unverdicted novelty 6.0 of 10

    RL-RIG uses a generate-reflect-edit loop with reinforcement learning to improve spatial accuracy in image generation, reporting up to 11% gains over prior open-source models on scene-graph metrics.

  6. Lavida-O: Elastic Large Masked Diffusion Models for Unified Multimodal Understanding and Generation

    cs.CV 2025-09 conditional novelty 6.0 of 10

    Lavida-O introduces an elastic mixture-of-transformers architecture that brings high-resolution text-to-image generation, object grounding, and image editing into a single masked diffusion model, using planning and se...

  7. Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning

    cs.CV 2026-06 unverdicted novelty 5.0 of 10

    A survey of test-time scaling for multimodal foundation models that introduces a three-way taxonomy of sampling, feedback, and search approaches along with applications and benchmarks.

Pith tools