Sidecar: Training-Free Semantic Reuse for Character-Consistent Free-form Visual Storytelling

• URL: http://arxiv.org/abs/2608.27280v1
• Date: Thu, 27 Aug 2026 15:53:30 GMT
• Title: Sidecar: Training-Free Semantic Reuse for Character-Consistent Free-form Visual Storytelling
• Abstract summary: In free-form story generation, a character is fully described only when first introduced and is later referred to by a type-level mention or pronoun.<n>We propose textbfSidecar, a plug-and-play semantic augmentation module that preserves entity-level information from the initial description and injects the missing semantics into later prompt embeddings.<n>Experiments on FreeStoryBench show that Sidecar consistently improves prompt-image alignment and character consistency across multiple SDXL- and FLUX-based baselines, with negligible computational overhead.
• Score: 5.249739038963074
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Visual storytelling requires generating images that follow a narrative while preserving consistent character identities across frames. In free-form story generation, a character is fully described only when first introduced and is later referred to by a type-level mention or pronoun. Although this setting better reflects natural storytelling, later prompts may omit important identity-related semantics, making character consistency more difficult to maintain. We propose \textbf{Sidecar}, a plug-and-play semantic augmentation module that preserves entity-level information from the initial description and injects the missing semantics into later prompt embeddings. Sidecar requires no additional training and does not modify the architecture of the base diffusion model. Experiments on FreeStoryBench show that Sidecar consistently improves prompt-image alignment and character consistency across multiple SDXL- and FLUX-based baselines, with negligible computational overhead.

Related papers

• FreeStory: Training-Free Character Consistency for Free-Form Visual Storytelling [4.671002796177002]
  FreeStory is a training-free framework that reformulates character consistency under free-form prompts as entity-grounded feature reuse.<n>Our method associates reference mentions with their corresponding character descriptions and combines dynamic character masks, correspondence-aware feature matching, key-value injection, and query blending.<n>Experiments show that FreeStory achieves state-of-the-art performance among training-free methods on structured benchmarks and stronger overall consistency over baselines under free-form prompts.
  arXiv  Detail & Related papers  (2026-06-23T18:37:31Z)
• Prompt-Anchored Vision-Text Distillation for Lifelong Person Re-identification [60.02530716177415]
  Lifelong person re-identification (LReID) aims to train a generalizable model with sequentially collected data.<n> Existing exemplar-free approaches largely rely on visual-only distillation or parameter regularization.<n>We propose Prompt-Anchored vision-text Distillation (PAD), an asymmetric vision-text framework for semantic alignment and cross-domain generalization.
  arXiv  Detail & Related papers  (2026-05-06T15:23:13Z)
• ReCap: Lightweight Referential Grounding for Coherent Story Visualization [11.022891519635834]
  ReCap is a lightweight consistency framework that improves character stability and visual fidelity without modifying the base diffusion backbone.<n>ReCap's CORE (COnditional frame REferencing) module treats anaphors, in our case pronouns, as visual anchors, activating only when characters are referred to by a pronoun and conditioning on the preceding frame to propagate visual identity.<n>We extend story visualization to human-centric narratives derived from real films, demonstrating the capability of ReCap beyond stylized cartoon domains.
  arXiv  Detail & Related papers  (2026-04-20T17:57:50Z)
• DeCorStory: Gram-Schmidt Prompt Embedding Decorrelation for Consistent Storytelling [1.7683026013361776]
  DeCorStory is a training-free inference-time framework that reduces inter-frame semantic interference.<n>It applies prompt embedding decorrelation to frame-level semantics, followed by singular value reweighting to strengthen prompt-specific information.<n> Experiments demonstrate consistent improvements in prompt-image alignment, identity consistency, and visual diversity.
  arXiv  Detail & Related papers  (2026-02-01T16:07:30Z)
• ReDiStory: Region-Disentangled Diffusion for Consistent Visual Story Generation [6.4611000755192585]
  ReDiStory is a training-free framework that improves multi-frame story generation via inference-time prompt embedding reorganization.<n>It reduces cross-frame interference without modifying diffusion parameters or requiring additional supervision.<n> Experiments on the ConsiStory+ benchmark show consistent gains over 1Prompt1Story on multiple identity consistency.
  arXiv  Detail & Related papers  (2026-02-01T16:04:40Z)
• StoryWeaver: A Unified World Model for Knowledge-Enhanced Story Character Customization [36.14275850149665]
  We propose a novel knowledge graph, namely Character Graph (textbfCG), which comprehensively represents various story-related knowledge.<n>We then introduce StoryWeaver, an image generator that achieve Customization via Character Graph (textbfC-CG), capable of consistent story visualization with rich text semantics.
  arXiv  Detail & Related papers  (2024-12-10T10:16:50Z)
• Evolving Storytelling: Benchmarks and Methods for New Character Customization with Diffusion Models [79.21968152209193]
  We introduce the NewEpisode benchmark to evaluate generative models' adaptability in generating new stories with fresh characters. We propose EpicEvo, a method that customizes a diffusion-based visual story generation model with a single story featuring the new characters seamlessly integrating them into established character dynamics.
  arXiv  Detail & Related papers  (2024-05-20T07:54:03Z)
• When StyleGAN Meets Stable Diffusion: a $\mathscr{W}_+$ Adapter for Personalized Image Generation [60.305112612629465]
  Text-to-image diffusion models have excelled in producing diverse, high-quality, and photo-realistic images. We present a novel use of the extended StyleGAN embedding space $mathcalW_+$ to achieve enhanced identity preservation and disentanglement for diffusion models. Our method adeptly generates personalized text-to-image outputs that are not only compatible with prompt descriptions but also amenable to common StyleGAN editing directions.
  arXiv  Detail & Related papers  (2023-11-29T09:05:14Z)
• Text-Only Training for Visual Storytelling [107.19873669536523]
  We formulate visual storytelling as a visual-conditioned story generation problem. We propose a text-only training method that separates the learning of cross-modality alignment and story generation.
  arXiv  Detail & Related papers  (2023-08-17T09:32:17Z)
• Make-A-Story: Visual Memory Conditioned Consistent Story Generation [57.691064030235985]
  We propose a novel autoregressive diffusion-based framework with a visual memory module that implicitly captures the actor and background context. Our method outperforms prior state-of-the-art in generating frames with high visual quality. Our experiments for story generation on the MUGEN, the PororoSV and the FlintstonesSV dataset show that our method not only outperforms prior state-of-the-art in generating frames with high visual quality, but also models appropriate correspondences between the characters and the background.
  arXiv  Detail & Related papers  (2022-11-23T21:38:51Z)
• StoryDALL-E: Adapting Pretrained Text-to-Image Transformers for Story Continuation [76.44802273236081]
  We develop a model StoryDALL-E for story continuation, where the generated visual story is conditioned on a source image. We show that our retro-fitting approach outperforms GAN-based models for story continuation and facilitates copying of visual elements from the source image. Overall, our work demonstrates that pretrained text-to-image synthesis models can be adapted for complex and low-resource tasks like story continuation.
  arXiv  Detail & Related papers  (2022-09-13T17:47:39Z)

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.