Think-Then-Generate: Reasoning-Aware Text-to-Image Diffusion with LLM Encoders

• URL: http://arxiv.org/abs/2601.10332v1
• Date: Thu, 15 Jan 2026 12:19:05 GMT
• Title: Think-Then-Generate: Reasoning-Aware Text-to-Image Diffusion with LLM Encoders
• Authors: Siqi Kou, Jiachun Jin, Zetong Zhou, Ye Ma, Yugang Wang, Quan Chen, Peng Jiang, Xiao Yang, Jun Zhu, Kai Yu, Zhijie Deng,
• Abstract summary: We propose a think-then-rewrite (T2G) paradigm for text-to-image (T2I) diffusion models.<n>We show substantial improvements in factual consistency, semantic alignment, and visual realism across reasoning-based image generation and editing benchmarks.<n>Our results constitute a promising step toward next-generation unified models with reasoning, expression, and demonstration capacities.
• Score: 46.79030733172859
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent progress in text-to-image (T2I) diffusion models (DMs) has enabled high-quality visual synthesis from diverse textual prompts. Yet, most existing T2I DMs, even those equipped with large language model (LLM)-based text encoders, remain text-pixel mappers -- they employ LLMs merely as text encoders, without leveraging their inherent reasoning capabilities to infer what should be visually depicted given the textual prompt. To move beyond such literal generation, we propose the think-then-generate (T2G) paradigm, where the LLM-based text encoder is encouraged to reason about and rewrite raw user prompts; the states of the rewritten prompts then serve as diffusion conditioning. To achieve this, we first activate the think-then-rewrite pattern of the LLM encoder with a lightweight supervised fine-tuning process. Subsequently, the LLM encoder and diffusion backbone are co-optimized to ensure faithful reasoning about the context and accurate rendering of the semantics via Dual-GRPO. In particular, the text encoder is reinforced using image-grounded rewards to infer and recall world knowledge, while the diffusion backbone is pushed to produce semantically consistent and visually coherent images. Experiments show substantial improvements in factual consistency, semantic alignment, and visual realism across reasoning-based image generation and editing benchmarks, achieving 0.79 on WISE score, nearly on par with GPT-4. Our results constitute a promising step toward next-generation unified models with reasoning, expression, and demonstration capacities.

Related papers

• Unified Multimodal Model as Auto-Encoder [69.38946823657592]
  We introduce a paradigm regarding understanding as the encoder (I2T) that compresses images into text, and generation as the decoder (T2I) that reconstructs images from that text.<n>Our empirical results suggest that understanding can largely enhance generation (verified on GenEval), while generation, in turn, notably strengthens fine-grained visual perception.
  arXiv  Detail & Related papers  (2025-09-11T17:57:59Z)
• A Comprehensive Study of Decoder-Only LLMs for Text-to-Image Generation [30.041283605038316]
  Many text-to-image models still employ the somewhat outdated T5 and CLIP as their text encoders.<n>We build a standardized training and evaluation pipeline that allows us to isolate and evaluate the effect of different text embeddings.<n>Our experiments reveal that the de facto way of using last-layer embeddings as conditioning leads to inferior performance.
  arXiv  Detail & Related papers  (2025-06-09T20:29:53Z)
• EDITOR: Effective and Interpretable Prompt Inversion for Text-to-Image Diffusion Models [41.63451923844824]
  We propose a prompt inversion technique called sys for text-to-image diffusion models.<n>Our method outperforms existing methods in terms of image similarity, textual alignment, prompt interpretability and generalizability.
  arXiv  Detail & Related papers  (2025-06-03T16:44:15Z)
• Decoder-Only LLMs are Better Controllers for Diffusion Models [63.22040456010123]
  We propose to enhance text-to-image diffusion models by borrowing the strength of semantic understanding from large language models.<n>Our adapter module is superior to the stat-of-the-art models in terms of text-to-image generation quality and reliability.
  arXiv  Detail & Related papers  (2025-02-06T12:17:35Z)
• Mimir: Improving Video Diffusion Models for Precise Text Understanding [53.72393225042688]
  Text serves as the key control signal in video generation due to its narrative nature.<n>The recent success of large language models (LLMs) showcases the power of decoder-only transformers.<n>This work addresses this challenge with Mimir, an end-to-end training framework featuring a carefully tailored token fuser.
  arXiv  Detail & Related papers  (2024-12-04T07:26:44Z)
• FLIER: Few-shot Language Image Models Embedded with Latent Representations [2.443383032451177]
  Few-shot Language Image model embedded with latent representations (FLIER) for image recognition. We first generate images and corresponding latent representations via Stable Diffusion with the textual inputs from GPT-3. With latent representations as "models-understandable pixels", we introduce a flexible convolutional neural network with two convolutional layers to be the latent encoder.
  arXiv  Detail & Related papers  (2024-10-10T06:27:46Z)
• LLM4GEN: Leveraging Semantic Representation of LLMs for Text-to-Image Generation [30.897935761304034]
  We propose a novel framework called textbfLLM4GEN, which enhances the semantic understanding of text-to-image diffusion models. A specially designed Cross-Adapter Module (CAM) integrates the original text features of text-to-image models with LLM features. DensePrompts, which contains $7,000$ dense prompts, provides a comprehensive evaluation for the text-to-image generation task.
  arXiv  Detail & Related papers  (2024-06-30T15:50:32Z)
• Free-Bloom: Zero-Shot Text-to-Video Generator with LLM Director and LDM Animator [59.589919015669274]
  This study focuses on zero-shot text-to-video generation considering the data- and cost-efficient. We propose a novel Free-Bloom pipeline that harnesses large language models (LLMs) as the director to generate a semantic-coherence prompt sequence. We also propose a series of annotative modifications to adapting LDMs in the reverse process, including joint noise sampling, step-aware attention shift, and dual-path.
  arXiv  Detail & Related papers  (2023-09-25T19:42:16Z)
• Text to Image Generation with Semantic-Spatial Aware GAN [41.73685713621705]
  A text to image generation (T2I) model aims to generate photo-realistic images which are semantically consistent with the text descriptions. We propose a novel framework Semantic-Spatial Aware GAN, which is trained in an end-to-end fashion so that the text encoder can exploit better text information.
  arXiv  Detail & Related papers  (2021-04-01T15:48:01Z)
• Enhanced Modality Transition for Image Captioning [51.72997126838352]
  We build a Modality Transition Module (MTM) to transfer visual features into semantic representations before forwarding them to the language model. During the training phase, the modality transition network is optimised by the proposed modality loss. Experiments have been conducted on the MS-COCO dataset demonstrating the effectiveness of the proposed framework.
  arXiv  Detail & Related papers  (2021-02-23T07:20:12Z)

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.