Aperture: Merge-Consistent Rotary States for Compressed Tokens
cs.AI
Submitted: 2026-09-29
Updated: 2026-09-29
Sources
- Qwen2.5-VL Technical Report
- Extending Context Window of Large Language Models via Positional Interpolation
- EchoingPixels: Aliasing-Resistant Joint Token Reduction for Audio-Visual LLMs
- PPE: Positional Preservation Embedding for Token Compression in Multimodal Large Language Models
- Time2Vec: Learning a Vector Representation of Time
- OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding?
- StreamingBench: Assessing the Gap for MLLMs to Achieve Streaming Video Understanding
- Enhancing Multi-Scale Implicit Learning in Image Super-Resolution with Integrated Positional Encoding
- LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models
- TIE: Time Interval Encoding for Video Generation over Events
- StreamingEval: A Unified Evaluation Protocol towards Realistic Streaming Video Understanding
- SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
- Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
- VideoRoPE: What Makes for Good Video Rotary Position Embedding?
- Qwen2.5-Omni Technical Report
- Apollo: An Exploration of Video Understanding in Large Multimodal Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems