# Better Tokens for Better 3D: Advancing Vision-Language Modeling in 3D Medical Imaging

arXiv: https://arxiv.org/abs/2510.20639 · HTML: https://mlsanity.com/paper/2510.20639/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 45 of its 62 references that are in the map.
- First posted on arXiv 23 Oct 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 30 of 45)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network](https://mlsanity.com/paper/1609.04802/) (2016)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [UNETR: Transformers for 3D Medical Image Segmentation](https://mlsanity.com/paper/2103.10504/) (2021)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers](https://mlsanity.com/paper/2401.08740/) (2024)
- [Simple Baselines for Image Restoration](https://mlsanity.com/paper/2204.04676/) (2022)
- [Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation](https://mlsanity.com/paper/2310.05737/) (2023)
- [MONAI: An open-source framework for deep learning in healthcare](https://mlsanity.com/paper/2211.02701/) (2022)
- [Generating Radiology Reports via Memory-driven Transformer](https://mlsanity.com/paper/2010.16056/) (2020)
- [InternVideo2: Scaling Foundation Models for Multimodal Video Understanding](https://mlsanity.com/paper/2403.15377/) (2024)
- [Towards Generalist Biomedical AI](https://mlsanity.com/paper/2307.14334/) (2023)
- [Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography](https://mlsanity.com/paper/2403.17834/) (2024)
- [Scaling Autoregressive Video Models](https://mlsanity.com/paper/1906.02634/) (2019)
- [Interactive and Explainable Region-guided Radiology Report Generation](https://mlsanity.com/paper/2304.08295/) (2023)
- [Make Pixels Dance: High-Dynamic Video Generation](https://mlsanity.com/paper/2311.10982/) (2023)
- [On the Challenges and Perspectives of Foundation Models for Medical Image Analysis](https://mlsanity.com/paper/2306.05705/) (2023)
- [Video Pixel Networks](https://mlsanity.com/paper/1610.00527/) (2016)
- [ChatGPT Makes Medicine Easy to Swallow: An Exploratory Case Study on Simplified Radiology Reports](https://mlsanity.com/paper/2212.14882/) (2022)
- [CT2Rep: Automated Radiology Report Generation for 3D Medical Imaging](https://mlsanity.com/paper/2403.06801/) (2024)

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

- [U-VLM: Hierarchical Vision Language Modeling for Report Generation](https://mlsanity.com/paper/2603.00479/) (2026)
- [CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference](https://mlsanity.com/paper/2604.06036/) (2026)
- [Beyond the Embedding Bottleneck: Adaptive Retrieval-Augmented 3D CT Report Generation](https://mlsanity.com/paper/2603.15822/) (2026)
- [Enhancing Fine-Grained Spatial Grounding in 3D CT Report Generation via Discriminative Guidance](https://mlsanity.com/paper/2604.10437/) (2026)
- [The Path to Self-Evolving Clinical Systems: Scaling Medical Agents from Assistance to Autonomy](https://mlsanity.com/paper/2607.11175/) (2026)
- [Foundation VAEs for 3D CT Reconstruction, Augmentation, and Generation](https://mlsanity.com/paper/2605.30893/) (2026)
- [BrainG3N: A Dual-Purpose Tokenizer for Controllable 3D Brain MRI Generation](https://mlsanity.com/paper/2606.19651/) (2026)
- [Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography](https://mlsanity.com/paper/2606.25546/) (2026)
- [CORTEX: A Structured Reasoning Benchmark for Trustworthy 3D Chest CT MLLMs](https://mlsanity.com/paper/2606.27264/) (2026)
- [RadSight: Towards Perceptually Reliable Multimodal Radiology Image Understanding](https://mlsanity.com/paper/2607.22293/) (2026)
- [When Do Cheap Probes Predict Expensive Training? Probing 3D-CT Encoders for Text Generation](https://mlsanity.com/paper/2607.22771/) (2026)
- [Rad-JEPA 3D: Radiology Joint-Embedding Predictive Model for 3D Computed Tomography](https://mlsanity.com/paper/2607.26196/) (2026)
- [Resolution Meets Reduction: Efficient Visual Context for 3D Radiology Report Generation](https://mlsanity.com/paper/2608.08713/) (2026)
- [AgentsEval: Clinically Faithful Evaluation of Medical Imaging Reports via Multi-Agent Reasoning](https://mlsanity.com/paper/2601.16685/) (2026)
- [EasyLens: A Training-Free Plug-and-Play Subtle-Lesion Representation Amplifier for Medical Vision-Language Models](https://mlsanity.com/paper/2606.06379/) (2026)
- [A Multi-Center Benchmark for Abdominal Disease Diagnosis and Report Generation from Non-Contrast CT](https://mlsanity.com/paper/2606.16991/) (2026)
- [Volumetric Radiology AI in the Era of Multimodal Large Language Models](https://mlsanity.com/paper/2608.20549/) (2026)
- [What Makes a Good Medical Image Tokenizer? Rethinking Reconstruction and Generation in Medical Image Tokenization](https://mlsanity.com/paper/2609.24691/) (2026)
- [NV-Reason-CT: 3D Visual Language Model for CT Analysis](https://mlsanity.com/paper/2609.27511/) (2026)
- [SelfCue: Making a 3D CT Report Generator Say What It Already Knows](https://mlsanity.com/paper/2609.31788/) (2026)

## Abstract (arXiv)

Recent progress in vision-language modeling for 3D medical imaging has been fueled by large-scale computed tomography (CT) corpora with paired free-text reports, stronger architectures, and powerful pretrained models. This has enabled applications such as automated report generation and text-conditioned 3D image synthesis. Yet, current approaches struggle with high-resolution, long-sequence volumes: contrastive pretraining often yields vision encoders that are misaligned with clinical language, and slice-wise tokenization blurs fine anatomy, reducing diagnostic performance on downstream tasks. We introduce BTB3D (Better Tokens for Better 3D), a causal convolutional encoder-decoder that unifies 2D and 3D training and inference while producing compact, frequency-aware volumetric tokens. A three-stage training curriculum enables (i) local reconstruction, (ii) overlapping-window tiling, and (iii) long-context decoder refinement, during which the model learns from short slice excerpts yet generalizes to scans exceeding 300 slices without additional memory overhead. BTB3D sets a new state-of-the-art on two key tasks: it improves BLEU scores and increases clinical F1 by 40% over CT2Rep, CT-CHAT, and Merlin for report generation; and it reduces FID by 75% and halves FVD compared to GenerateCT and MedSyn for text-to-CT synthesis, producing anatomically consistent 512*512*241 volumes. These results confirm that precise three-dimensional tokenization, rather than larger language backbones alone, is essential for scalable vision-language modeling in 3D medical imaging. The codebase is available at: https://github.com/ibrahimethemhamamci/BTB3D