# What Makes for Good Visual Tokenizers for Large Language Models?

arXiv: https://arxiv.org/abs/2305.12223 · HTML: https://mlsanity.com/paper/2305.12223/

## Key facts

- Cited by 42 papers in the map (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 41 of its 54 references that are in the map.
- First posted on arXiv 20 May 2023.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)

## Builds on (most cited first, 30 of 41)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Exploring Simple Siamese Representation Learning](https://mlsanity.com/paper/2011.10566/) (2020)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Semantic Understanding of Scenes through the ADE20K Dataset](https://mlsanity.com/paper/1608.05442/) (2016)
- [mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality](https://mlsanity.com/paper/2304.14178/) (2023)

## Built on by (most cited first, 30 of 42)

- [SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension](https://mlsanity.com/paper/2307.16125/) (2023)
- [V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs](https://mlsanity.com/paper/2312.14135/) (2023)
- [LVLM-eHub: A Comprehensive Evaluation Benchmark for Large Vision-Language Models](https://mlsanity.com/paper/2306.09265/) (2023)
- [UniTok: A Unified Tokenizer for Visual Generation and Understanding](https://mlsanity.com/paper/2502.20321/) (2025)
- [SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension](https://mlsanity.com/paper/2404.16790/) (2024)
- [SEED-Bench-2: Benchmarking Multimodal Large Language Models](https://mlsanity.com/paper/2311.17092/) (2023)
- [CapsFusion: Rethinking Image-Text Data at Scale](https://mlsanity.com/paper/2310.20550/) (2023)
- [A Survey on Benchmarks of Multimodal Large Language Models](https://mlsanity.com/paper/2408.08632/) (2024)
- [LION : Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge](https://mlsanity.com/paper/2311.11860/) (2023)
- [EgoPlan-Bench: Benchmarking Multimodal Large Language Models for Human-Level Planning](https://mlsanity.com/paper/2312.06722/) (2023)
- [UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognition](https://mlsanity.com/paper/2311.15599/) (2023)
- [Pink: Unveiling the Power of Referential Comprehension for Multi-modal LLMs](https://mlsanity.com/paper/2310.00582/) (2023)
- [Visual Representation Alignment for Multimodal Large Language Models](https://mlsanity.com/paper/2509.07979/) (2025)
- [Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models](https://mlsanity.com/paper/2503.17349/) (2025)
- [How Far Are We From AGI: Are LLMs All We Need?](https://mlsanity.com/paper/2405.10313/) (2024)
- [A Survey on Multimodal Benchmarks: In the Era of Large AI Models](https://mlsanity.com/paper/2409.18142/) (2024)
- [MouSi: Poly-Visual-Expert Vision-Language Models](https://mlsanity.com/paper/2401.17221/) (2024)
- [A Comprehensive Evaluation of GPT-4V on Knowledge-Intensive Visual Question Answering](https://mlsanity.com/paper/2311.07536/) (2023)
- [MMDocBench: Benchmarking Large Vision-Language Models for Fine-Grained Visual Document Understanding](https://mlsanity.com/paper/2410.21311/) (2024)
- [AVTrustBench: Assessing and Enhancing Reliability and Robustness in Audio-Visual LLMs](https://mlsanity.com/paper/2501.02135/) (2025)
- [Unified Generative and Discriminative Training for Multi-modal Large Language Models](https://mlsanity.com/paper/2411.00304/) (2024)
- [X-Former: Unifying Contrastive and Reconstruction Learning for MLLMs](https://mlsanity.com/paper/2407.13851/) (2024)
- [ReForm-Eval: Evaluating Large Vision Language Models via Unified Re-Formulation of Task-Oriented Benchmarks](https://mlsanity.com/paper/2310.02569/) (2023)
- [TinyLVLM-eHub: Towards Comprehensive and Efficient Evaluation for Large Vision-Language Models](https://mlsanity.com/paper/2308.03729/) (2023)
- [Cyclic Contrastive Knowledge Transfer for Open-Vocabulary Object Detection](https://mlsanity.com/paper/2503.11005/) (2025)
- [Exploring and Evaluating Multimodal Knowledge Reasoning Consistency of Multimodal Large Language Models](https://mlsanity.com/paper/2503.04801/) (2025)
- [4D-Bench: Benchmarking Multi-modal Large Language Models for 4D Object Understanding](https://mlsanity.com/paper/2503.17827/) (2025)
- [Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities](https://mlsanity.com/paper/2401.14405/) (2024)
- [AgriBench: A Hierarchical Agriculture Benchmark for Multimodal Large Language Models](https://mlsanity.com/paper/2412.00465/) (2024)
- [VTBench: Evaluating Visual Tokenizers for Autoregressive Image Generation](https://mlsanity.com/paper/2505.13439/) (2025)

## Abstract (arXiv)

We empirically investigate proper pre-training methods to build good visual tokenizers, making Large Language Models (LLMs) powerful Multimodal Large Language Models (MLLMs). In our benchmark, which is curated to evaluate MLLMs visual semantic understanding and fine-grained perception capabilities, we discussed different visual tokenizers pre-trained with dominant methods (i.e., DeiT, CLIP, MAE, DINO), and observe that: i) Fully/weakly supervised models capture more semantics than self-supervised models, but the gap is narrowed by scaling up the pre-training dataset. ii) Self-supervised models are better at fine-grained perception, where patch-level supervision is particularly effective. iii) Tuning the visual tokenizer leads to the loss of semantics obtained from large-scale pretraining, which is unfavorable with relatively small-scale instruction-tuning dataset. Given the findings, we reviewed methods that attempted to unify semantics and fine-grained visual understanding, e.g., patch-level feature distillation with semantically-rich targets. We obtain an intriguing insight mask-based strategies that were once all the rage may not be applicable for obtaining good visual tokenizers. Based on this critical observation, we obtain a new MLLM equipped with a tailored Good Visual Tokenizer (GVT), which exhibits strong visual comprehension capability at multiple scales. In particular, without introducing extra parameters and task-specific fine-tuning, GVT achieves superior performance on visual question answering, image captioning, and other fine-grained visual understanding tasks such as object counting and multi-class identification.