# Stepwise Token Selection for Efficient Multimodal Large Language Models

arXiv: https://arxiv.org/abs/2606.16067 · HTML: https://mlsanity.com/paper/2606.16067/

## Key facts

- Cited by 5 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 37 of its 43 references that are in the map.
- First posted on arXiv 14 Jun 2026.

## 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 37)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [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)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [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)
- [Pointer Networks](https://mlsanity.com/paper/1506.03134/) (2015)
- [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)
- [LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models](https://mlsanity.com/paper/2403.15388/) (2024)
- [SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference](https://mlsanity.com/paper/2410.04417/) (2024)
- [VisionZip: Longer is Better but Not Necessary in Vision Language Models](https://mlsanity.com/paper/2412.04467/) (2024)
- [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)
- [ATP-LLaVA: Adaptive Token Pruning for Large Vision Language Models](https://mlsanity.com/paper/2412.00447/) (2024)
- [TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model](https://mlsanity.com/paper/2503.18278/) (2025)
- [Don't Just Chase "Highlighted Tokens" in MLLMs: Revisiting Visual Holistic Context Retention](https://mlsanity.com/paper/2510.02912/) (2025)
- [Segmentation and Vascular Vectorization for Coronary Artery by Geometry-based Cascaded Neural Network](https://mlsanity.com/paper/2305.04208/) (2023)
- [A Glimpse to Compress: Dynamic Visual Token Pruning for Large Vision-Language Models](https://mlsanity.com/paper/2508.01548/) (2025)
- [MedViLaM: A multimodal large language model with advanced generalizability and explainability for medical data understanding and generation](https://mlsanity.com/paper/2409.19684/) (2024)
- [TC-SSA: Token Compression via Semantic Slot Aggregation for Gigapixel Pathology Reasoning](https://mlsanity.com/paper/2603.01143/) (2026)
- [One Leaf Reveals the Season: Occlusion-Based Contrastive Learning with Semantic-Aware Views for Efficient Visual Representation](https://mlsanity.com/paper/2411.09858/) (2024)
- [p-MoD: Building Mixture-of-Depths MLLMs via Progressive Ratio Decay](https://mlsanity.com/paper/2412.04449/) (2024)
- [Fewer Tokens, Greater Scaling: Self-Adaptive Visual Bases for Efficient and Expansive Representation Learning](https://mlsanity.com/paper/2511.19515/) (2025)
- [A foundation model for generalizable disease diagnosis in chest X-ray images](https://mlsanity.com/paper/2410.08861/) (2024)

## Built on by (most cited first, 5 of 5)

- [PathSelect: Sequential Token Selection for Whole Slide Pathology](https://mlsanity.com/paper/2607.23631/) (2026)
- [Decoupling semantics from vision: A framework for faithful visual-text compression evaluation](https://mlsanity.com/paper/2608.01848/) (2026)
- [DiffPrune: differentiable information throttling for token pruning in vision-language models](https://mlsanity.com/paper/2608.01985/) (2026)
- [StepPrune: Adaptive Sequential Visual Token Selection across Multimodal Large Language Models](https://mlsanity.com/paper/2609.13804/) (2026)
- [CoViST: Visual Token Compression via Composable States](https://mlsanity.com/paper/2609.33397/) (2026)

## Abstract (arXiv)

In multimodal large language models (MLLMs), inference cost is largely dominated by the visual token prefix rather than the language backbone, making token reduction a key factor for improving efficiency. Existing approaches typically assign independent importance scores to visual tokens and retain a fixed number of top-ranked tokens, implicitly assuming token independence and a uniform compression ratio across inputs. In this work, we reformulate visual token pruning as a sequential decision-making process. Specifically, we introduce a pointer-style selection mechanism that iteratively chooses informative tokens, conditioning each decision on previously selected ones, and dynamically determines when to stop via a learned termination action. This enables joint optimization of both the selected subset and its size. To enable end-to-end training under standard language modeling objectives, we design a differentiable relaxation based on a variance-preserving noise interpolation scheme, allowing gradients to propagate through the discrete selection process. Extensive experiments on LLaVA-v1.5-7B and Qwen2.5-VL-7B demonstrate that our approach consistently outperforms fixed-ratio baselines across different compression levels. Under aggressive pruning that removes 88.9% of visual tokens, our method preserves 94.6% of the original accuracy while achieving a 1.88x speed-up in prefill latency.