# TrimTokenator: Towards Adaptive Visual Token Pruning for Large Multimodal Models

arXiv: https://arxiv.org/abs/2509.00320 · HTML: https://mlsanity.com/paper/2509.00320/

## Key facts

- Cited by 9 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 21 of its 63 references that are in the map.
- First posted on arXiv 30 Aug 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, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [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)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge](https://mlsanity.com/paper/1906.00067/) (2019)
- [Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet](https://mlsanity.com/paper/2101.11986/) (2021)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [nocaps: novel object captioning at scale](https://mlsanity.com/paper/1812.08658/) (2018)
- [AdaViT: Adaptive Tokens for Efficient Vision Transformer](https://mlsanity.com/paper/2112.07658/) (2021)
- [PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction](https://mlsanity.com/paper/2410.17247/) (2024)
- [On The Computational Complexity of Self-Attention](https://mlsanity.com/paper/2209.04881/) (2022)
- [Boosting Multimodal Large Language Models with Visual Tokens Withdrawal for Rapid Inference](https://mlsanity.com/paper/2405.05803/) (2024)
- [Q-Transformer: Scalable Offline Reinforcement Learning via Autoregressive Q-Functions](https://mlsanity.com/paper/2309.10150/) (2023)
- [Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs](https://mlsanity.com/paper/2506.10967/) (2025)
- [FlowCut: Rethinking Redundancy via Information Flow for Efficient Vision-Language Models](https://mlsanity.com/paper/2505.19536/) (2025)
- [CATP: Contextually Adaptive Token Pruning for Efficient and Enhanced Multimodal In-Context Learning](https://mlsanity.com/paper/2508.07871/) (2025)
- [Multi-Granular Multimodal Clue Fusion for Meme Understanding](https://mlsanity.com/paper/2503.12560/) (2025)
- [PDTrim: Targeted Pruning for Prefill-Decode Disaggregation in Inference](https://mlsanity.com/paper/2509.04467/) (2025)

## Built on by (most cited first, 9 of 9)

- [VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning](https://mlsanity.com/paper/2601.15724/) (2026)
- [HIVTP: A Training-Free Method to Improve VLMs Efficiency via Hierarchical Visual Token Pruning Using Middle-Layer-Based Importance Score](https://mlsanity.com/paper/2509.23663/) (2025)
- [How Many Visual Tokens Do Multimodal Language Models Need? Scaling Visual Token Pruning with F^3A](https://mlsanity.com/paper/2605.16359/) (2026)
- [Combating Textual Noise and Redundancy: Entropy-Aware Dense Visual Token Pruning](https://mlsanity.com/paper/2607.02484/) (2026)
- [When Does Visual Token Pruning Improve Calibration? The Role of Evidence Coverage in MLLMs](https://mlsanity.com/paper/2604.12035/) (2026)
- [MedEvoEval: Evaluating Continual Evolution of Doctor Agents through Simulated Clinical Episodes](https://mlsanity.com/paper/2606.28900/) (2026)
- [FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization](https://mlsanity.com/paper/2606.28933/) (2026)
- [EVLA: An Electro-Aware Multimodal Assistant for Physically-Grounded Driving Reasoning and Control](https://mlsanity.com/paper/2606.28938/) (2026)
- [A3M: Adaptive, Adversarial and Multi-Objective Learning for Strategic Bidding in Repeated Auctions](https://mlsanity.com/paper/2606.28943/) (2026)

## Abstract (arXiv)

Large Multimodal Models (LMMs) have achieved significant success across various tasks. These models usually encode visual inputs into dense token sequences, which are then concatenated with textual tokens and jointly processed by a language model. However, the increased token count substantially raises computational and memory costs during inference. Token pruning has emerged as a promising approach to address this issue. Existing token pruning methods often rely on costly calibration or suboptimal importance metrics, leading to redundant retained tokens. In this paper, we analyze the redundancy differences between visual and textual tokens and propose pruning exclusively on visual tokens. Based on this, we propose a visual token pruning strategy that explicitly preserves both cross-modal alignment and intra-modal informational diversity. We introduce a mutual information-based token pruning strategy that removes visual tokens semantically misaligned with textual tokens, effectively preserving the alignment between the visual and textual modalities. To further improve the representational quality of the retained tokens, we additionally prune redundant visual tokens by maximizing the expected pairwise distances in the embedding space, which is solved efficiently with a greedy algorithm. Extensive experiments demonstrate that our method maintains strong performance while reducing tokens by 88.9% on models such as LLaVA-1.5-7B and LLaVA-NEXT-7B, resulting in a 56.7% improvement in inference speed.