Stage-adaptive Token Selection for Efficient Omni-modal LLMs

arXiv:2605.20035 · cs.CV · Submitted 2026-05-19 · Read on arXiv

cs.CV

Submitted: 2026-05-19

Updated: 2026-09-27

Code: https://github.com/xxayt/SEATS

Terminology

• Pruning — 22× in this paper · explained in 4 episodes
• Benchmark — 12× in this paper · explained in 2 episodes
• Redundancy — 11× in this paper · explained in 2 episodes
• Accuracy — 5× in this paper · explained in 1 episode
• Ablation — 3× in this paper · explained in 1 episode
• Ablation Studies — 3× in this paper · explained in 1 episode
• Alignment — 2× in this paper · explained in 1 episode
• Computational overhead — 2× in this paper · explained in 2 episodes
• Embeddings — 2× in this paper · explained in 2 episodes
• Robustness — 2× in this paper · explained in 5 episodes
• Calibration — 1× in this paper · explained in 7 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Core — 1× in this paper · explained in 2 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Decoupling — 1× in this paper · explained in 2 episodes
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• KV cache — 1× in this paper · explained in 3 episodes

Sources

• Qwen3-VL Technical Report
• OmniSIFT: Modality-Asymmetric Token Compression for Efficient Omni-modal Large Language Models
• ARC-Hunyuan-Video-7B: Structured Video Comprehension of Real-World Shorts
• EchoingPixels: Aliasing-Resistant Joint Token Reduction for Audio-Visual LLMs
• LLaVA-OneVision: Easy Visual Task Transfer
• LLaVA-NeXT-Interleave: Tackling Multi-image, Video, and 3D in Large Multimodal Models
• Baichuan-Omni-1.5 Technical Report
• video-SALMONN 2: Caption-Enhanced Audio-Visual Large Language Models
• LVOmniBench: Pioneering Long Audio-Video Understanding Evaluation for Omnimodal LLMs
• Qwen3.5-Omni Technical Report
• InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
• Mini-Omni2: Towards Open-source GPT-4o with Vision, Speech and Duplex Capabilities
• Qwen2.5-Omni Technical Report
• Qwen3-Omni Technical Report
• MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe
• Daily-Omni: Towards Audio-Visual Reasoning with Temporal Alignment across Modalities

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models