rotary position embedding

Anchoring and Rescaling Attention for Semantically Coherent Inbetweening CVPR 2026 AToken: A Unified Tokenizer for Vision CVPR 2026 Extending Audio Context for Long-Form Understanding in Large Audio-Language Models EACL 2026 GenHOI: Towards Object-Consistent Hand-Object Interaction with Temporally Balanced and Spatially Selective Object Injection CVPR 2026 Infinity-RoPE: Action-Controllable Infinite Video Generation Emerges From Autoregressive Self-Rollout CVPR 2026 LoL: Longer than Longer, Scaling Video Generation to Hour CVPR 2026 MMFace-DiT: A Dual-Stream Diffusion Transformer for High-Fidelity Multimodal Face Generation CVPR 2026 MultiShotMaster: A Controllable Multi-Shot Video Generation Framework CVPR 2026 NAF: Zero-Shot Feature Upsampling via Neighborhood Attention Filtering CVPR 2026 PAS: A Training-Free Stabilizer for Temporal Encoding in Video LLMs CVPR 2026 ReDirector: Creating Any-Length Video Retakes with Rotary Camera Encoding CVPR 2026 RoSA: Enhancing Parameter-Efficient Fine-Tuning via RoPE-aware Selective Adaptation in Large Language Models AAAI 2026 Towards Long-window Anchoring in Vision-Language Model Distillation AAAI 2026 Advancing General Multimodal Capability of Vision-language Models with Pyramid-descent Visual Position Encoding ACL 2025 Beyond Position: the emergence of wavelet-like properties in Transformers ACL 2025 Breaking the Stage Barrier: A Novel Single-Stage Approach to Long Context Extension for Large Language Models COLING 2025 CommVQ: Commutative Vector Quantization for KV Cache Compression ICML 2025 ComRoPE: Scalable and Robust Rotary Position Embedding Parameterized by Trainable Commuting Angle Matrices CVPR 2025 DynImg: Key Frames with Visual Prompts are Good Representation for Multi-Modal Video Understanding ICCV 2025 Extending LLM Context Window with Adaptive Grouped Positional Encoding: A Training-Free Method ACL 2025 Fourier Position Embedding: Enhancing Attention’s Periodic Extension for Length Generalization ICML 2025 FreeFlux: Understanding and Exploiting Layer-Specific Roles in RoPE-Based MMDiT for Versatile Image Editing ICCV 2025 MASH-VLM: Mitigating Action-Scene Hallucination in Video-LLMs through Disentangled Spatial-Temporal Representations CVPR 2025 PEPE: Long-context Extension for Large Language Models via Periodic Extrapolation Positional Encodings EMNLP 2025 RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations IJCAI 2025