LogiStory: A Logic-Aware Framework for Multi-Image Story Visualization

Dimensions overall score 7.0

• LogiStory significantly improves narrative logic in generated visual stories compared to existing methods.result 90%
  Evidencepartial

  Experiments demonstrate that our approach significantly improves the narrative logic of generated visual stories.

  Implicationpartial

  Directly stated in abstract with supporting quantitative results in evaluation tables showing LogiStory achieves highest scores in Narrative Causality and Story Readability.

  Verificationpartial

  partial

• Existing multimodal models struggle with maintaining logical flow in visual sequence generation, resulting in disjointed actions and fragmented narratives.limitation 85%
  Evidencepartial

  existing models still struggle to maintain logical flow, often resulting in disjointed actions, fragmented narratives, and unclear storylines.

  Implicationpartial

  Explicitly stated in abstract as the problem motivation, with clear description of current limitations.

  Verificationpartial

  partial

• LogiStory uses a multi-agent system that explicitly models visual logic by grounding roles, extracting causal chains, and verifying story-level consistency.method 88%
  Evidencepartial

  we design a multi-agent system that grounds roles, extracts causal chains, and verifies story-level consistency, transforming narrative coherence from an implicit byproduct of image generation into an explicit modeling objective.

  Implicationpartial

  Directly described in abstract as the core methodological innovation, with specific components mentioned.

  Verificationpartial

  partial

• The LogicTale benchmark provides explicit causal logic annotations and supports multi-image joint evaluation, unlike prior datasets.technical 87%
  Evidencepartial

  LogicTale provides explicit causal logic annotations and supports multi-image joint evaluation, enabling systematic assessment of narrative coherence.

  Implicationpartial

  Explicitly stated in Table 7 comparison with clear differentiation from existing benchmarks.

  Verificationpartial

  partial

• LogiStory achieves the highest scores in Narrative Causality, Story Readability, Aesthetic Quality and Character Expressiveness among compared methods.result 92%
  Evidencepartial

  Our approach achieves the highest score in Narrative Causality, Story Readability, Aesthetic Quality and Character Expressiveness.

  Implicationpartial

  Direct quantitative claim supported by results in Figure 3/Table 1 showing LogiStory outperforming all listed baselines.

  Verificationpartial

  partial

• Visual logic is defined as the perceptual and causal coherence among characters, actions, and scenes over time in visual sequence generation.technical 90%
  Evidencepartial

  visual logic, a critical yet underexplored dimension of visual sequence generation that we define as the perceptual and causal coherence among characters, actions, and scenes over time.

  Implicationpartial

  Explicit definition provided in abstract as a core conceptual contribution.

  Verificationpartial

  partial

• LogiStory excels in narrative coherence and visual storytelling compared to baselines that may achieve slightly higher scores on low-level quality metrics.result 82%
  Evidencepartial

  LogiStory excels in narrative coherence and visual storytelling, which aligns with the central goals of our framework.

  Implicationpartial

  Directly stated in analysis section with acknowledgment of trade-offs, though slightly inferential from comparative language.

  Verificationpartial

  partial

Competitive landscape

LogiStory is a framework that explicitly models visual logic to generate coherent and communicative multi-image stories, addressing a key limitation in current multimodal systems.