Preprint

RADIO1D: Elastic Representations for Condensed Vision Modeling

Greg Heinrich Michael Ranzinger Collin McCarthy N. Bagrov Eugene Khvedchenya Bryan Catanzaro Jan Kautz Andrew Tao Pavlo Molchanov

Jul 2026 · 0 citations · 48 references

Computer Science

TL;DR

RADIO1D is introduced, which compresses images into a compact, variable-length 1D token sequence using multi-teacher knowledge distillation and an autoencoder design, delivering competitive performance on diverse multimodal benchmarks with lower computational overhead and better accuracy.

Abstract

This paper challenges the assumption that vision-language models (VLMs) require fixed patch-based 2D vision features. Analyzing fine-tuned vision encoders, we find that representations become increasingly abstract and less spatially coherent during VLM training. Notably, models trained with image-text alignment (such as SigLIP2) develop a small number of specialized tokens that effectively summarize global image content. Building on this, we introduce RADIO1D, which compresses images into a compact, variable-length 1D token sequence using multi-teacher knowledge distillation and an autoencoder design. The resulting representations exhibit strong hierarchical summarization, enabling accurate scene understanding - even with a single token - and support improved composition-aware image retrieval. In VLMs, RADIO1D provides flexible accuracy-efficiency tradeoffs through adjustable token counts, delivering competitive performance on diverse multimodal benchmarks with lower computational overhead and better accuracy.

arXiv Semantic Scholar

View source

Similar papers

Jul 2026

MAViE: A Multi-scale Adaptive Vision Encoder for Fine-grained Visual Perception and Efficient Multimodal Reasoning

This work introduces \method, a Multi-scale Adaptive Vision Encoder, a Multi-scale Adaptive Vision Encoder that uses position-dependent gates to fuse shallow, intermediate, and deep features from a vision Transformer, preserving global semantics while enhancing edges, text, and local structure.

Sha Lei · arXiv.org · 0 citations

Preprint Aug 2026

MLLMCLIP: Feature-Level Distillation of MLLM for Robust Vision-Language Representations

Compared to prior CLIP-enhancement methods, MLLMCLIP achieves state-of-the-art compositional accuracy while delivering consistent gains on standard zero-shot classification and image-text retrieval, showing that feature-level distillation strengthens both compositional and general vision-language representation capability.

Jongsuk Kim, Qiyu Wu, Zhuoyuan Mao et al. · 0 citations

Preprint Aug 2026

MoE-ViE: Mixture of Experts Vision Encoder for Efficient Image and Video Understanding

This work systematically study MoE designs for vision encoder scaling and finds that fine-grained MoE topologies yield substantial gains over both dense and standard MoE counterparts, and proposes an auxiliary-loss-free balancing variant for better expert utilization, and designs a specialized MoE kernel to mitigate inference latency overhead.

Bonan Zhang, Shiyu Dong, Quan Hung Tran et al. · 0 citations

Preprint Aug 2026

HAFI-VLM: A Frequency Perspective for Diagnosing and Enhancing Visual Perception in Vision-Language Models

Mechanistic analyses show that HAFI restores task-dependent spectral allocation while retaining semantic attention, establishing frequency enrichment as a distinct and effective route for improving VLM perception.

Jin Cui, Chuanchang Su, Jia-Yi Lu et al. · 0 citations

Preprint Jul 2026

SigLIP-HD by Fine-to-Coarse Supervision

This work studies an interesting problem: how to achieve fine visual perception under lower cost without larger images, and builds this framework on the advanced SigLIP 2 model, which consistently delivers stronger results than the baseline model, especially on OCR-related tasks.

Lihe Yang, Zhen Zhao, Hengshuang Zhao · 0 citations

Preprint Aug 2026

Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding

Qwen-3D is introduced, a geometry-aware LMM that compresses visual information within the Qwen backbone using multi-view geometric cues, enabling efficient long-horizon visual reasoning over static scenes and incorporates a query-based segmentation decoder that grounds language directly in the underlying 3D scene representation.

Lucy Lin, Ayush Jain, Yifan Liu et al. · 0 citations