REVIEW 7 cited by
Scalable Image Tokenization with Index Backpropagation Quantization
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Existing vector quantization (VQ) methods struggle with scalability, largely attributed to the instability of the codebook that undergoes partial updates during training. The codebook is prone to collapse as utilization decreases, due to the progressively widening distribution gap between non-activated codes and visual features. To solve the problem, we propose Index Backpropagation Quantization (IBQ), a new VQ method for the joint optimization of all codebook embeddings and the visual encoder. Applying a straight-through estimator on the one-hot categorical distribution between the encoded feature and codebook, all codes are differentiable and maintain a consistent latent space with the visual encoder. IBQ enables scalable training of visual tokenizers and, for the first time, achieves a large-scale codebook ($2^{18}$) with high dimension ($256$) and high utilization. Experiments on the standard ImageNet benchmark demonstrate the scalability and superiority of IBQ, achieving competitive results on reconstruction and the application of autoregressive visual generation. The code and models are available at https://github.com/TencentARC/SEED-Voken.
Forward citations
Cited by 7 Pith papers
-
dRAE: Representation Autoencoder with Hyper-Spherical Codes
Switching codebook assignment and update to cosine similarity while keeping a magnitude-preserving commitment loss avoids codebook collapse and scales visual tokenizers to 131,072 codes with high utilization.
-
Emu3.5: Native Multimodal Models are World Learners
Emu3.5 is a native multimodal world model pre-trained on over 10 trillion vision-language tokens with next-token prediction, post-trained via reinforcement learning, and accelerated by Discrete Diffusion Adaptation fo...
-
Scalable Training for Vector-Quantized Networks with 100% Codebook Utilization
A training-only ViT-based projector, VQBridge, combined with learning annealing, achieves full codebook utilization in vector-quantized networks at large codebook sizes, improving reconstruction and autoregressive ima...
-
Kronos: A Foundation Model for the Language of Financial Markets
Kronos, a foundation model for financial K-line data using discrete tokenization and autoregressive training, raises price-forecasting RankIC by about 90% over general time-series foundation models.
-
Rethinking Discrete Tokens: Treating Them as Conditions for Continuous Autoregressive Image Synthesis
DisCon treats discrete image tokens as conditioning signals rather than targets, letting a continuous autoregressive model refine details and reach gFID 1.38 on ImageNet-256.
-
HumanGenesis: Agent-Based Geometric and Generative Modeling for Synthetic Human Dynamics
HumanGenesis is a claimed state-of-the-art framework that couples 3D Gaussian reconstruction, LLM-based critique, pose guidance, and diffusion-based harmonization for synthetic human video generation; unverified in th...
-
VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers
A convolutional residual VQ-VAE action tokenizer trained on over 100x more data than prior work improves OpenVLA success rates and inference speed on several manipulation tasks.
Discussion (0). Sign in to comment.