text-encoder

Tag

Cards List
#text-encoder

Which Negatives Matter? Ask Your Text Encoder: Adaptive Similarity Margins for Dense-Caption Retrieval

arXiv cs.AI · 2026-08-20 Cached

The paper introduces HN-CLIP, a method that uses the text encoder's text-text geometry to create adaptive similarity margins for dense-caption retrieval, addressing saturation issues in contrastive learning and improving performance over existing methods.

0 favorites 0 likes
#text-encoder

sakamakismile/Qwen3-VL-32B-Heretic-MiniMax-H3-NVFP4

Hugging Face Models Trending · 2026-08-04 Cached

Release of an NVFP4-quantized uncensored MiniMax-H3 text encoder (Qwen3-VL-32B Heretic) that fits on a single 16GB GPU and serves as a drop-in replacement in ComfyUI workflows.

0 favorites 0 likes
#text-encoder

UR-BERT: Scaling Text Encoders for Massively Multilingual TTS Through Universal Romanization and Speech Token Prediction

arXiv cs.CL · 2026-06-11 Cached

UR-BERT proposes a Romanized transcription-based text encoder for massively multilingual TTS, scaling to 495 languages by using universal Romanization and a speech token prediction objective to enhance phonetic alignment and generalization to unseen languages.

0 favorites 0 likes
#text-encoder

Comfy-Org/Ideogram-4

Hugging Face Models Trending · 2026-06-03 Cached

Ideogram-4 model repackaged for ComfyUI, including fp8 scaled diffusion models, Qwen3VL text encoder, and FLUX VAE.

0 favorites 0 likes
#text-encoder

Text-to-Image Models Need Less from Text Encoders Than You Think

Hugging Face Daily Papers · 2026-06-02 Cached

This paper demonstrates that text-to-image diffusion transformer models primarily rely on token merging and word order from text encoders rather than full contextual embeddings, suggesting that the image model itself decodes complex linguistic structures.

0 favorites 0 likes
#text-encoder

Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models

arXiv cs.CL · 2026-04-20 Cached

This paper investigates how semantic information is distributed across textual tokens in text-to-image models, finding that information concentration and cross-item interactions significantly affect image generation alignment. The authors use patching techniques to demonstrate that simple encoding-stage interventions can improve alignment quality.

0 favorites 0 likes
← Back to home

Submit Feedback