# CoVeR: Coverage-Based Token Pruning for Multi-View 3D Reasoning in VLMs

arXiv: https://arxiv.org/abs/2609.08345 · HTML: https://mlsanity.com/paper/2609.08345/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Visual token pruning in multimodal models, one of 12 directions in Efficient LLM inference and serving.
- Builds on 72 of its 81 references that are in the map.
- First posted on arXiv 8 Sep 2026.

## Placed in

- 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 72)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes](https://mlsanity.com/paper/1702.04405/) (2017)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs](https://mlsanity.com/paper/2406.07476/) (2024)
- [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)
- [Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI](https://mlsanity.com/paper/2109.08238/) (2021)
- [LLaVA-Video: Video Instruction Tuning With Synthetic Data](https://mlsanity.com/paper/2410.02713/) (2024)
- [MovieChat: From Dense Token to Sparse Memory for Long Video Understanding](https://mlsanity.com/paper/2307.16449/) (2023)
- [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)
- [Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding](https://mlsanity.com/paper/2311.08046/) (2023)
- [An Embodied Generalist Agent in 3D World](https://mlsanity.com/paper/2311.12871/) (2023)
- [ScanQA: 3D Question Answering for Spatial Scene Understanding](https://mlsanity.com/paper/2112.10482/) (2021)
- [Deep Modular Co-Attention Networks for Visual Question Answering](https://mlsanity.com/paper/1906.10770/) (2019)
- [PointLLM: Empowering Large Language Models to Understand Point Clouds](https://mlsanity.com/paper/2308.16911/) (2023)
- [SQA3D: Situated Question Answering in 3D Scenes](https://mlsanity.com/paper/2210.07474/) (2022)

## Abstract (arXiv)

Representing a 3D scene as multi-view images allows 2D VLMs to reason in 3D by reusing priors from pre-training, sidestepping the scarcity of annotated 3D data. However, it produces thousands of redundant visual tokens whose cost grows with every view. Existing visual token pruners fall into two families, each limited in the 3D multi-view setting. Learned importance methods rank tokens by attention or encoder features; because redundancy here is fundamentally spatial, they keep near-duplicate tokens from a few prominent regions and leave most of the scene unrepresented. Voxelization methods improve spatial coverage but cannot enforce an exact token budget and saturate as multi-view observations overlap in 3D, capping retention well below the target. We show that spatial coverage is associated with 3D reasoning performance and introduce CoVeR, a deterministic, training-free selector that uses only token coordinates, with no learned signals. CoVeR selects tokens that collectively cover every region of the scene, and solves the limitations of both families: it enforces an exact per-scene budget, breaks the voxelization saturation plateau, and avoids the near-duplicate selections of learned importance. Extensive experiments show CoVeR outperforms prior SOTAs on all three 3D reasoning benchmarks and generalizes as a plug-and-play module tested across four VLMs. Notably, with only ≈8% of visual tokens, it preserves 93.5% of full-token performance, surpassing SOTA by 3.9 percentage points on average across benchmarks.