A technique for controllable diversity in text-to-image generation by inducing structured semantic variations at the prompt level via VLM and agentic workflow.
Generating an image from 1,000 words: Enhancing text-to-image with structured captions
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 7roles
background 1polarities
support 1representative citing papers
A lightweight Q-Former proxy trained on VLM hidden states reveals that localization signals peak in input-dependent intermediate layers, not the final layers used by standard editing pipelines.
ReflectCAP distills model-specific hallucination and oversight patterns into Structured Reflection Notes that steer LVLMs toward more factual and complete image captions, reaching the Pareto frontier on factuality-coverage trade-offs.
LingBot-Video is an open-source MoE video foundation model for embodied intelligence that scales to 120B parameters, integrates robot data, and uses multi-dimensional RL to improve physical plausibility.
APE post-trains small language models as single-agent or multi-agent prompt enhancers that improve visual alignment on image generation and editing benchmarks without altering the downstream visual model.
LTX-2 generates high-quality synchronized audiovisual content from text prompts via an asymmetric 14B-video / 5B-audio dual-stream transformer with cross-attention and modality-aware guidance.
Token-to-Token alignment rephrases prompts into shared structure then matches token embeddings by semantic similarity, making linear interpolation a meaningful operation for blending in text-to-image models.
citing papers explorer
-
Semantic Browsing: Controllable Diversity for Image Generation
A technique for controllable diversity in text-to-image generation by inducing structured semantic variations at the prompt level via VLM and agentic workflow.
-
Analysis-by-Proxy: Localization Signals in VLMs Operating as Condition Encoders
A lightweight Q-Former proxy trained on VLM hidden states reveals that localization signals peak in input-dependent intermediate layers, not the final layers used by standard editing pipelines.
-
ReflectCAP: Detailed Image Captioning with Reflective Memory
ReflectCAP distills model-specific hallucination and oversight patterns into Structured Reflection Notes that steer LVLMs toward more factual and complete image captions, reaching the Pareto frontier on factuality-coverage trade-offs.
-
Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence
LingBot-Video is an open-source MoE video foundation model for embodied intelligence that scales to 120B parameters, integrates robot data, and uses multi-dimensional RL to improve physical plausibility.
-
APE: Agentic Prompt Enhancer for Image Generation and Editing
APE post-trains small language models as single-agent or multi-agent prompt enhancers that improve visual alignment on image generation and editing benchmarks without altering the downstream visual model.
-
LTX-2: Efficient Joint Audio-Visual Foundation Model
LTX-2 generates high-quality synchronized audiovisual content from text prompts via an asymmetric 14B-video / 5B-audio dual-stream transformer with cross-attention and modality-aware guidance.
-
Token-to-Token Alignment of Text Embeddings for Semantic Blending
Token-to-Token alignment rephrases prompts into shared structure then matches token embeddings by semantic similarity, making linear interpolation a meaningful operation for blending in text-to-image models.