AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Discrete Tokenization for Multimodal LLMs: A Comprehensive Survey

Jindong Li, Yali Fu, Jiahong Liu, L. J. Cao, Wei Ji, Menglin Yang, Irwin King, Ming–Hsuan Yang

TPAMI 2026 · Large language models › Language, safety and interpretability · approach: LLM pruning and scaling data · 16 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use FSQ instead of VQGAN quantizerImage tokenizer · ImageNet generation 128x128gFID: 8.01 -> 7.33Improved
use BSQ instead of VQGAN quantizerImage tokenizer · ImageNet reconstruction 256x256rFID: 1.228 -> 0.700Improved
use FSQ instead of VQGAN quantizerImage tokenizer · ImageNet generation 128x128IS: 151.40 -> 224.88Improved
use LFQ instead of VQGAN quantizerImage tokenizer · ImageNet reconstruction 256x256rFID: 1.228 -> 2.444Hurt
use LFQ instead of VQGAN quantizerImage tokenizer · ImageNet generation 128x128gFID: 8.01 -> 7.73Improved
use RVQ/CQ/AQ multi-codebook full-dimensional quantization instead of PQ subspace splittingANN search · SIFT1M ANN search (64-bit codes)Recall@1: 0.224 -> 0.298Improved
use AQ instead of PQANN search · SIFT1M ANN search (64-bit codes)Recall@10: 0.599 -> 0.741Improved
use AQ instead of PQANN search · SIFT1M ANN search (64-bit codes)Recall@100: 0.924 -> 0.972Improved
use RVQ instead of PQANN search · SIFT1M ANN search (64-bit codes)Recall@1: 0.224 -> 0.259Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy