InsightTok: Improving Text and Face Fidelity in Discrete Tokenization for Autoregressive Image Generation

Hugging Face Daily Papers Papers
discrete-tokenization autoregressive-image-generation text-fidelity face-fidelity perceptual-losses visual-tokenization

Summary

InsightTok introduces content-aware perceptual losses to improve discrete visual tokenization for better text and face reconstruction, enhancing autoregressive image generation quality.

Text and faces are among the most perceptually salient and practically important patterns in visual generation, yet they remain challenging for autoregressive generators built on discrete tokenization. A central bottleneck is the tokenizer: aggressive downsampling and quantization often discard the fine-grained structures needed to preserve readable glyphs and distinctive facial features. We attribute this gap to standard discrete-tokenizer objectives being weakly aligned with text legibility and facial fidelity, as these objectives typically optimize generic reconstruction while compressing diverse content uniformly. To address this, we propose InsightTok, a simple yet effective discrete visual tokenization framework that enhances text and face fidelity through localized, content-aware perceptual losses. With a compact 16k codebook and a 16x downsampling rate, InsightTok significantly outperforms prior tokenizers in text and face reconstruction without compromising general reconstruction quality. These gains consistently transfer to autoregressive image generation in InsightAR, producing images with clearer text and more faithful facial details. Overall, our results highlight the potential of specialized supervision in tokenizer training for advancing discrete image generation.
Original Article
View Cached Full Text

Cached at: 05/18/26, 02:23 AM

Paper page - InsightTok: Improving Text and Face Fidelity in Discrete Tokenization for Autoregressive Image Generation

Source: https://huggingface.co/papers/2605.14333 Published on May 14

#2 Paper of the day Authors:

,

,

,

,

,

,

,

,

,

,

,

Abstract

InsightTok improves discrete visual tokenization for better text and face reconstruction through content-aware perceptual losses, enhancing autoregressive image generation quality.

Text and faces are among the most perceptually salient and practically important patterns in visual generation, yet they remain challenging forautoregressive generatorsbuilt on discrete tokenization. A central bottleneck is thetokenizer: aggressivedownsamplingand quantization often discard the fine-grained structures needed to preserve readable glyphs and distinctive facial features. We attribute this gap to standarddiscrete-tokenizer objectivesbeing weakly aligned withtext legibilityandfacial fidelity, as these objectives typically optimize generic reconstruction while compressing diverse content uniformly. To address this, we propose InsightTok, a simple yet effectivediscrete visual tokenizationframework that enhances text and face fidelity through localized, content-awareperceptual losses. With a compact 16kcodebookand a 16xdownsamplingrate, InsightTok significantly outperforms priortokenizers in text and face reconstruction without compromising general reconstruction quality. These gains consistently transfer toautoregressive image generationin InsightAR, producing images with clearer text and more faithful facial details. Overall, our results highlight the potential of specialized supervision intokenizertraining for advancing discrete image generation.

View arXiv pageView PDFGitHub9Add to collection

Get this paper in your agent:

hf papers read 2605\.14333

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.14333 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.14333 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.14333 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

(1D) Ordered Tokens Enable Efficient Test-Time Search

Hugging Face Daily Papers

This paper investigates how 1D coarse-to-fine token structures in autoregressive models improve test-time search efficiency compared to classical 2D grid tokenization. The authors show that such ordered tokens enable better test-time scaling and even training-free text-to-image generation when guided by image-text verifiers.

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

arXiv cs.CL

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.

TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment

Hugging Face Daily Papers

TIPSv2 introduces enhanced vision-language pretraining techniques including patch-level distillation, an upgraded masked image objective (iBOT++), and improved caption sampling strategies to achieve superior dense patch-text alignment. The resulting family of image-text encoder models demonstrates strong performance across 9 tasks and 20 datasets.

Stochasticity in Tokenization Improves Robustness

arXiv cs.CL

This paper demonstrates that training large language models with stochastic tokenization instead of deterministic canonical tokenization significantly improves robustness to adversarial attacks and random perturbations, with improvements shown across pre-training, fine-tuning, and in-context learning without increasing inference costs.