Tag
Parallel Tube Decoding enables efficient simultaneous spatial and temporal video grounding by eliminating autoregressive dependencies, reducing latency and improving accuracy over standard methods.
CORA-Diff is a training-free method that accelerates diffusion language model inference by using native confidence and persistence signals to accept residual positions early, skipping redundant dense denoising passes while preserving task quality.
Introduces Archer, a training-free KV caching method for diffusion language models that adaptively reuses cached hidden states to reduce recomputation while preserving rollback capabilities, achieving up to 2.95x speedup and improved generation quality.
This paper proposes GenCDSR, a generative framework for cross-domain sequential recommendation with hybrid tokenization and serial-parallel decoding, achieving improved accuracy and significantly reduced inference latency compared to state-of-the-art baselines.
MUGEN introduces a unified motion-language framework that avoids discrete codebooks and iterative decoding, using a single adaptive-length autoencoder with continuous latent slots and one-shot generation to achieve efficient, high-quality text-to-motion and motion-to-text performance across HumanML3D and SnapMoGen benchmarks.
DIRECT is a framework for sequence labeling using large language models that improves domain alignment through Direct Preference Optimization (DPO) after supervised fine-tuning and increases inference efficiency via controlled decoding with template-filling and KV cache reuse.
NVIDIA releases Nemotron-Labs-Diffusion, a family of tri-mode language models (3B, 8B, 14B) supporting AR, diffusion, and self-speculation decoding, achieving 2.7x-4x speed-ups over standard AR decoding.