Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs

arXiv:2609.03820 · cs.CV, cs.CL · Submitted 2026-09-03 · Read on arXiv

cs.CV, cs.CL

Submitted: 2026-09-03

Updated: 2026-09-03

Code: https://github.com/codeprakhar25/omp-keyframe-sampling

Project page: https://vision.cs.utexas.edu/projects/adaalloc

Terminology

• Benchmark — 32× in this paper · explained in 2 episodes
• Accuracy — 30× in this paper · explained in 1 episode
• Embeddings — 4× in this paper · explained in 2 episodes
• Modality Gap — 3× in this paper · explained in 4 episodes
• Pruning — 3× in this paper · explained in 4 episodes
• Trace — 3× in this paper · explained in 2 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Adaptation — 2× in this paper · explained in 1 episode
• Allocator — 2× in this paper · explained in 1 episode
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes

Sources

• LDDR: Linear-DPP-Based Dynamic-Resolution Frame Sampling for Video MLLMs
• Event-Anchored Frame Selection for Effective Long-Video Understanding
• LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models
• AdaCodec: A Predictive Visual Code for Video MLLMs
• Adaptive Greedy Frame Selection for Long Video Understanding
• ReQuest: Rethinking-based Question-Aware Frame Selection for Long-Form Video QA
• Q-Frame: Query-aware Frame Selection and Multi-Resolution Adaptation for Video-LLMs
• QCA: Query- and Content-Aware Keyframe Selection for Long Video Understanding
• AdaRD-key: Adaptive Relevance-Diversity Keyframe Sampling for Long-form Video understanding
• AdaptToken: Entropy-based Adaptive Token Selection for MLLM Long Video Understanding
• InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
• Qwen3-VL Technical Report
• FOCUS: Efficient Keyframe Selection for Long Video Understanding
• MDP3: A Training-free Approach for List-wise Frame Selection in Video-LLMs
• Adaptive Keyframe Sampling for Long Video Understanding
• LVBench: An Extreme Long Video Understanding Benchmark
• Query-Conditioned Evidential Keyframe Sampling for MLLM-Based Long-Form Video Understanding
• Efficient Frame Selection for Long Videos at Test Time with Attention-Based MLLM Selectors
• LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding

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