Salt++: Context-Aligned Post-Training for Few-Step Streaming Multimodal Generation

arXiv:2609.36995 · cs.CV, cs.AI, eess.IV · Submitted 2026-09-29 · Read on arXiv

cs.CV, cs.AI, eess.IV

Submitted: 2026-09-29

Updated: 2026-09-29

Project page: https://xingtongge.github.io/Saltpp

Terminology

• Distillation — 48× in this paper · explained in 2 episodes
• Consistency — 18× in this paper · explained in 3 episodes
• Alignment — 17× in this paper · explained in 1 episode
• Causal Self-Flow — 12× in this paper
• Distribution Matching — 11× in this paper · explained in 2 episodes
• Ablation — 8× in this paper · explained in 1 episode
• Adaptation — 7× in this paper · explained in 1 episode
• Flow Matching — 5× in this paper · explained in 8 episodes
• Information Asymmetry — 5× in this paper · explained in 2 episodes
• Consistency Distillation — 4× in this paper · explained in 2 episodes
• Diffusion Model — 4× in this paper · explained in 8 episodes
• Diffusion Models — 4× in this paper · explained in 6 episodes
• Distribution Matching Distillation — 3× in this paper
• Foundation Model — 3× in this paper · explained in 3 episodes
• Generative Modeling — 3× in this paper · explained in 3 episodes
• Representation Learning — 3× in this paper · explained in 2 episodes
• Transformer — 3× in this paper · explained in 2 episodes
• Ablation Studies — 2× in this paper · explained in 1 episode
• Calibration — 2× in this paper · explained in 7 episodes
• KV cache — 2× in this paper · explained in 3 episodes
• Provenance — 2× in this paper · explained in 3 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Alignment Scores — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Causal Inference — 1× in this paper · explained in 4 episodes
• Core — 1× in this paper · explained in 2 episodes
• Diffusion Process — 1× in this paper · explained in 2 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• Latent Space — 1× in this paper · explained in 7 episodes
• Optimal Transport — 1× in this paper · explained in 5 episodes
• Rectified Flow — 1× in this paper · explained in 3 episodes
• Semantic Alignment — 1× in this paper · explained in 3 episodes
• World Model — 1× in this paper · explained in 7 episodes

Sources

• Context-Matched Distillation: Teacher Causality for Autoregressive Video Distillation
• Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
• Speed by Simplicity: A Single-Stream Architecture for Fast Audio-Video Generative Foundation Model
• Seedance 1.0: Exploring the Boundaries of Video Generation Models
• Salt: Self-Consistent Distribution Matching with Cache-Aware Training for Fast Video Generation
• LTX-2: Efficient Joint Audio-Visual Foundation Model
• Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion
• HunyuanVideo: A Systematic Framework For Large Video Generative Models
• Autoregressive Adversarial Post-Training for Real-Time Interactive Video Generation
• JavisDiT: Joint Audio-Video Diffusion Transformer with Hierarchical Spatio-Temporal Prior Synchronization
• Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
• Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation
• Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference
• Seedance 1.5 pro: A Native Audio-Visual Joint Generation Foundation Model
• Seedance 2.0: Advancing Video Generation for World Complexity
• OmniForcing: Unleashing Real-time Joint Audio-Visual Generation
• Kling-Omni Technical Report
• MOVA: Towards Scalable and Synchronized Video-Audio Generation
• MAGI-1: Autoregressive Video Generation at Scale
• Wan: Open and Advanced Large-Scale Video Generative Models

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