# HiRED: Attention-Guided Token Dropping for Efficient Inference of High-Resolution Vision-Language Models

arXiv: https://arxiv.org/abs/2408.10945 · HTML: https://mlsanity.com/paper/2408.10945/

## Key facts

- Cited by 28 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 19 of its 32 references that are in the map.
- First posted on arXiv 20 Aug 2024.

## 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, 19 of 19)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering](https://mlsanity.com/paper/2209.09513/) (2022)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://mlsanity.com/paper/2203.10244/) (2022)
- [DocVQA: A Dataset for VQA on Document Images](https://mlsanity.com/paper/2007.00398/) (2020)
- [Vision Transformers Need Registers](https://mlsanity.com/paper/2309.16588/) (2023)
- [LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models](https://mlsanity.com/paper/2403.15388/) (2024)
- [Monkey: Image Resolution and Text Label Are Important Things for Large Multi-modal Models](https://mlsanity.com/paper/2311.06607/) (2023)
- [LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models](https://mlsanity.com/paper/2407.12772/) (2024)
- [MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training](https://mlsanity.com/paper/2403.09611/) (2024)
- [mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understanding](https://mlsanity.com/paper/2403.12895/) (2024)
- [TinyLLaVA: A Framework of Small-scale Large Multimodal Models](https://mlsanity.com/paper/2402.14289/) (2024)
- [InternLM-XComposer2-4KHD: A Pioneering Large Vision-Language Model Handling Resolutions from 336 Pixels to 4K HD](https://mlsanity.com/paper/2404.06512/) (2024)
- [IA-RED2: Interpretability-Aware Redundancy Reduction for Vision Transformers](https://mlsanity.com/paper/2106.12620/) (2021)
- [Token-level Correlation-guided Compression for Efficient Multimodal Document Understanding](https://mlsanity.com/paper/2407.14439/) (2024)
- [Advancing High Resolution Vision-Language Models in Biomedicine](https://mlsanity.com/paper/2406.09454/) (2024)

## Built on by (most cited first, 28 of 28)

- [PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction](https://mlsanity.com/paper/2410.17247/) (2024)
- [Stop Looking for Important Tokens in Multimodal Language Models: Duplication Matters More](https://mlsanity.com/paper/2502.11494/) (2025)
- [See What You Are Told: Visual Attention Sink in Large Multimodal Models](https://mlsanity.com/paper/2503.03321/) (2025)
- [Multi-Stage Vision Token Dropping: Towards Efficient Multimodal Large Language Model](https://mlsanity.com/paper/2411.10803/) (2024)
- [Dynamic-LLaVA: Efficient Multimodal Large Language Models via Dynamic Vision-language Context Sparsification](https://mlsanity.com/paper/2412.00876/) (2024)
- [PACT: Pruning and Clustering-Based Token Reduction for Faster Visual Language Models](https://mlsanity.com/paper/2504.08966/) (2025)
- [Don't Just Chase "Highlighted Tokens" in MLLMs: Revisiting Visual Holistic Context Retention](https://mlsanity.com/paper/2510.02912/) (2025)
- [ZipVL: Efficient Large Vision-Language Models with Dynamic Token Sparsification](https://mlsanity.com/paper/2410.08584/) (2024)
- [VL-Cache: Sparsity and Modality-Aware KV Cache Compression for Vision-Language Model Inference Acceleration](https://mlsanity.com/paper/2410.23317/) (2024)
- [Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Acceleration](https://mlsanity.com/paper/2412.13180/) (2024)
- [VScan: Rethinking Visual Token Reduction for Efficient Large Vision-Language Models](https://mlsanity.com/paper/2505.22654/) (2025)
- [Variation-aware Vision Token Dropping for Faster Large Vision-Language Models](https://mlsanity.com/paper/2509.01552/) (2025)
- [FocusLLaVA: A Coarse-to-Fine Approach for Efficient and Effective Visual Token Compression](https://mlsanity.com/paper/2411.14228/) (2024)
- [VisiPruner: Decoding Discontinuous Cross-Modal Dynamics for Efficient Multimodal LLMs](https://mlsanity.com/paper/2510.17205/) (2025)
- [Frequency-Aware Autoregressive Modeling for Efficient High-Resolution Image Synthesis](https://mlsanity.com/paper/2507.20454/) (2025)
- [SparseVILA: Decoupling Visual Sparsity for Efficient VLM Inference](https://mlsanity.com/paper/2510.17777/) (2025)
- [Why 1 + 1 < 1 in Visual Token Pruning: Beyond Naive Integration via Multi-Objective Balanced Covering](https://mlsanity.com/paper/2505.10118/) (2025)
- [ECG-Byte: A Tokenizer for End-to-End Generative Electrocardiogram Language Modeling](https://mlsanity.com/paper/2412.14373/) (2024)
- [AuroraLong: Bringing RNNs Back to Efficient Open-Ended Video Understanding](https://mlsanity.com/paper/2507.02591/) (2025)
- [Connector-S: A Survey of Connectors in Multi-modal Large Language Models](https://mlsanity.com/paper/2502.11453/) (2025)
- [LightVLM: Acceleraing Large Multimodal Models with Pyramid Token Merging and KV Cache Compression](https://mlsanity.com/paper/2509.00419/) (2025)
- [LLaVA-Zip: Adaptive Visual Token Compression with Intrinsic Image Information](https://mlsanity.com/paper/2412.08771/) (2024)
- [ToFu: Visual Tokens Reduction via Fusion for Multi-modal, Multi-patch, Multi-image Task](https://mlsanity.com/paper/2503.04444/) (2025)
- [Multi-Cue Adaptive Visual Token Pruning for Large Vision-Language Models](https://mlsanity.com/paper/2503.08019/) (2025)
- [GreedyPrune: Retenting Critical Visual Token Set for Large Vision Language Models](https://mlsanity.com/paper/2506.13166/) (2025)
- [ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs](https://mlsanity.com/paper/2507.21420/) (2025)
- [Hierarchical Adaptive Eviction for KV Cache Management in Multimodal Language Models](https://mlsanity.com/paper/2602.02197/) (2026)
- [FoveaTree: Information-Density Trees for Query-Guided Visual Token Pruning in Video Large Language Models](https://mlsanity.com/paper/2610.16744/) (2026)

## Abstract (arXiv)

High-resolution Vision-Language Models (VLMs) are widely used in multimodal tasks to enhance accuracy by preserving detailed image information. However, these models often generate an excessive number of visual tokens due to the need to encode multiple partitions of a high-resolution image input. Processing such a large number of visual tokens through multiple transformer networks poses significant computational challenges, particularly for resource-constrained commodity GPUs. To address this challenge, we propose High-Resolution Early Dropping (HiRED), a plug-and-play token-dropping method designed to operate within a fixed token budget. HiRED leverages the attention of CLS token in the vision transformer (ViT) to assess the visual content of the image partitions and allocate an optimal token budget for each partition accordingly. The most informative visual tokens from each partition within the allocated budget are then selected and passed to the subsequent Large Language Model (LLM). We showed that HiRED achieves superior accuracy and performance, compared to existing token-dropping methods. Empirically, HiRED-20% (i.e., a 20% token budget) on LLaVA-Next-7B achieves a 4.7x increase in token generation throughput, reduces response latency by 78%, and saves 14% of GPU memory for single inference on an NVIDIA TESLA P40 (24 GB). For larger batch sizes (e.g., 4), HiRED-20% prevents out-of-memory errors by cutting memory usage by 30%, while preserving throughput and latency benefits. Code - https://github.com/hasanar1f/HiRED