FLARE is a vision-language model family using text-guided vision encoding, context-aware alignment decoding, dual-semantic mapping loss, and text-driven VQA synthesis to achieve deep cross-modal integration, outperforming larger models with only 630 vision tokens at 3B scale.
From pixels to prose: A large dataset of dense image cap- tions
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CV 6roles
dataset 1polarities
use dataset 1representative citing papers
Janus decouples visual encoding into task-specific pathways inside a single autoregressive transformer to unify multimodal understanding and generation while outperforming earlier unified models.
VCap pairs reference captions as witnesses with visual signals as adjudicators to deliver hypergeometric-precision rewards for RL in visual captioning, enabling an 8B model to outperform SOTA on benchmarks and improve weak-to-strong generalization.
BEiTScore is a new efficient cross-encoder metric for reference-free image captioning evaluation that achieves state-of-the-art results on detailed caption benchmarks through VQA initialization and adversarial LLM augmentations.
WinTok is a hybrid visual tokenizer that supplements pixel tokens with learnable semantic tokens distilled asymmetrically from foundation models to improve reconstruction, understanding, and generation.
DeepSeek-VL2 is a series of MoE vision-language models using dynamic tiling and latent attention that reach competitive or state-of-the-art results on VQA, OCR, document understanding and grounding with 1.0B to 4.5B activated parameters.
citing papers explorer
-
FLARE: Fully Integration of Vision-Language Representations for Deep Cross-Modal Understanding
FLARE is a vision-language model family using text-guided vision encoding, context-aware alignment decoding, dual-semantic mapping loss, and text-driven VQA synthesis to achieve deep cross-modal integration, outperforming larger models with only 630 vision tokens at 3B scale.
-
Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation
Janus decouples visual encoding into task-specific pathways inside a single autoregressive transformer to unify multimodal understanding and generation while outperforming earlier unified models.
-
VCap: Hypergeometric Rewards for Weak-to-Strong Visual Captioning
VCap pairs reference captions as witnesses with visual signals as adjudicators to deliver hypergeometric-precision rewards for RL in visual captioning, enabling an 8B model to outperform SOTA on benchmarks and improve weak-to-strong generalization.
-
BEiTScore: Reference-free Image Captioning Evaluation with an Efficient Cross-Encoder Model
BEiTScore is a new efficient cross-encoder metric for reference-free image captioning evaluation that achieves state-of-the-art results on detailed caption benchmarks through VQA initialization and adversarial LLM augmentations.
-
WinTok: A Win-Win Hybrid Tokenizer via Decomposing Visual Understanding and Generation with Transferable Tokens
WinTok is a hybrid visual tokenizer that supplements pixel tokens with learnable semantic tokens distilled asymmetrically from foundation models to improve reconstruction, understanding, and generation.