# Hybrid Token Compression for Vision-Language Models

arXiv: https://arxiv.org/abs/2512.08240 · HTML: https://mlsanity.com/paper/2512.08240/

## Key facts

- Cited by 20 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 53 of its 65 references that are in the map.
- First posted on arXiv 9 Dec 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)
- Efficient LLM inference and serving / [Visual token pruning in multimodal models](https://mlsanity.com/inference/visual-token-pruning-in-multimodal-models/)

## Builds on (most cited first, 30 of 53)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models](https://mlsanity.com/paper/2306.13394/) (2023)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [Transformers in Vision: A Survey](https://mlsanity.com/paper/2101.01169/) (2021)
- [SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension](https://mlsanity.com/paper/2307.16125/) (2023)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [Vision-Language Models for Vision Tasks: A Survey](https://mlsanity.com/paper/2304.00685/) (2023)
- [An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models](https://mlsanity.com/paper/2403.06764/) (2024)
- [LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models](https://mlsanity.com/paper/2311.17043/) (2023)
- [Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations](https://mlsanity.com/paper/2202.07800/) (2022)
- [Visual Transformers: Token-based Image Representation and Processing for Computer Vision](https://mlsanity.com/paper/2006.03677/) (2020)

## Built on by (most cited first, 20 of 20)

- [FastVMT: Eliminating Redundancy in Video Motion Transfer](https://mlsanity.com/paper/2602.05551/) (2026)
- [FlashVLM: Text-Guided Visual Token Selection for Large Multimodal Models](https://mlsanity.com/paper/2512.20561/) (2025)
- [STORM: Search-Guided Generative World Models for Robotic Manipulation](https://mlsanity.com/paper/2512.18477/) (2025)
- [CoAgent: Collaborative Planning and Consistency Agent for Coherent Video Generation](https://mlsanity.com/paper/2512.22536/) (2025)
- [Large Language Models as Discounted Bayesian Filters](https://mlsanity.com/paper/2512.18489/) (2025)
- [Adaptive-VoCo: Complexity-Aware Visual Token Compression for Vision-Language Models](https://mlsanity.com/paper/2512.18496/) (2025)
- [AgriWorld:A World Tools Protocol Framework for Verifiable Agricultural Reasoning with Code-Executing LLM Agents](https://mlsanity.com/paper/2602.15325/) (2026)
- [Reflective Confidence: Correcting Reasoning Flaws via Online Self-Correction](https://mlsanity.com/paper/2512.18605/) (2025)
- [ForgetMark: Stealthy Fingerprint Embedding via Targeted Unlearning in Language Models](https://mlsanity.com/paper/2601.08189/) (2026)
- [PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding](https://mlsanity.com/paper/2605.30126/) (2026)
- [GTMA: Dynamic Representation Optimization for OOD Vision-Language Models](https://mlsanity.com/paper/2512.18504/) (2025)
- [PTTA: A Pure Text-to-Animation Framework for High-Quality Creation](https://mlsanity.com/paper/2512.18614/) (2025)
- [SirenPose: Dynamic Scene Reconstruction via Geometric Supervision](https://mlsanity.com/paper/2512.20531/) (2025)
- [A Three-Level Alignment Framework for Large-Scale 3D Retrieval and Controlled 4D Generation](https://mlsanity.com/paper/2512.22294/) (2025)
- [Learning Dynamic Scene Reconstruction with Sinusoidal Geometric Priors](https://mlsanity.com/paper/2512.22295/) (2025)
- [Self-Rewarded Multimodal Coherent Reasoning Across Diverse Visual Domains](https://mlsanity.com/paper/2512.22545/) (2025)
- [Multi-Agent Cooperative Learning for Robust Vision-Language Alignment under OOD Concepts](https://mlsanity.com/paper/2601.09746/) (2026)
- [LogicGaze: Benchmarking Causal Consistency in Visual Narratives via Counterfactual Verification](https://mlsanity.com/paper/2602.00292/) (2026)
- [On the Sparsity-Storage-Accuracy Tradeoff in Parsimoniously Activated Dictionary Learning](https://mlsanity.com/paper/2606.22352/) (2026)
- [Query Independent Variable Rate Visual Token Coding](https://mlsanity.com/paper/2610.00204/) (2026)

## Abstract (arXiv)

Vision-language models (VLMs) rely on hundreds of visual tokens, leading to high computational and memory costs. Existing compression methods face a trade-off: continuous compression can weaken high-level semantics, while discrete quantization may lose fine-grained appearance details. We introduce HTC-VLM, a hybrid visual token compression framework that disentangles semantics and appearance through two complementary pathways. A continuous pathway preserves detailed ViT patch features, while a discrete pathway provides semantic anchors using MGVQ quantization represented by four tokens. The two pathways are fused into a 580-token hybrid sequence and compressed into a single token using a disentanglement attention mask and a <voco> bottleneck. Under the same one-token output budget, HTC-VLM achieves 87.2% average performance retention across seven benchmarks (GQA, VQAv2, MMBench, MME, POPE, SEED-Bench, and ScienceQA-Image), outperforming the leading continuous baseline at 81.0%. Attention analysis shows that the compressed token prioritizes discrete anchors, supporting their role as semantic guidance. We further study token-budget scaling, cross-architecture generalization, inference efficiency, robustness to codebook and masking variations, and the information-theoretic properties of the hybrid bottleneck. These results show that combining continuous appearance features with discrete semantic anchors enables effective extreme visual token compression for efficient VLMs.