Related papers: LongLive-2.0: An NVFP4 Parallel Infrastructure for…

Vision-language models (VLMs) could power real-time assistants and autonomous agents, but they face a critical challenge: understanding near-infinite video streams without escalating latency and memory usage. Processing entire videos with…

In recent years, Large Language Models (LLMs) have exhibited remarkable capabilities, driving advancements in real-world applications. However, training LLMs on increasingly long input sequences imposes significant challenges due to high…

Video generation is a critical pathway toward world models, with efficient long video inference as a key capability. Toward this end, we introduce LongCat-Video, a foundational video generation model with 13.6B parameters, delivering strong…

Despite rapid progress in autoregressive video diffusion, an emerging system algorithm bottleneck limits both deployability and generation capability: KV cache memory. In autoregressive video generation models, the KV cache grows with…

In long-context large language model (LLM) inference, the prefill stage dominates computation due to self-attention over the complete input context. Sparse attention significantly reduces self-attention computation by limiting each token's…

Video diffusion models (DMs) have enabled high-quality video synthesis. However, their computation costs scale quadratically with sequence length because self-attention has quadratic complexity. While linear attention lowers the cost, fully…

Sparse convolution plays a pivotal role in emerging workloads, including point cloud processing in AR/VR, autonomous driving, and graph understanding in recommendation systems. Since the computation pattern is sparse and irregular,…

Diffusion models have emerged as the prevailing approach for text-to-image (T2I) and text-to-video (T2V) generation, yet production platforms must increasingly serve both modalities on shared GPU clusters while meeting stringent latency…

Autoregressive video world models predict future visual observations conditioned on actions. While effective over short horizons, these models often struggle with long-horizon generation, as small prediction errors accumulate over time.…

Vision-Language Models (VLMs) are increasingly tasked with ultra-long multimodal understanding. While linear architectures offer constant computation and memory footprints, they often struggle with high-frequency visual perception compared…

Training large transformers is slow, but recent innovations on GPU architecture give us an advantage. NVIDIA Ampere GPUs can execute a fine-grained 2:4 sparse matrix multiplication twice as fast as its dense equivalent. In the light of this…

Long-video understanding has emerged as a crucial capability in real-world applications such as video surveillance, meeting summarization, educational lecture analysis, and sports broadcasting. However, it remains computationally…

Autoregressive (AR) video diffusion models adopt a streaming generation framework, enabling long-horizon video generation with real-time responsiveness, as exemplified by the Self Forcing training paradigm. However, existing AR video…

Linear sequence modeling approaches, such as linear attention, provide advantages like linear-time training and constant-memory inference over sequence lengths. However, existing sequence parallelism (SP) methods are either not optimized…

Diffusion models have revolutionized image and video generation, achieving unprecedented visual quality. However, their reliance on transformer architectures incurs prohibitively high computational costs, particularly when extending…

Streaming video large language models (LLMs) are increasingly used for real-time multimodal tasks such as video captioning, question answering, conversational agents, and augmented reality. However, these models face fundamental memory and…

Vision-language models (VLMs) predominantly rely on autoregressive decoding, which generates tokens one at a time and fundamentally limits inference throughput. This limitation is especially acute in physical AI scenarios such as robotics…

Human beings are endowed with a complementary learning system, which bridges the slow learning of general world dynamics with fast storage of episodic memory from a new experience. Previous video generation models, however, primarily focus…

Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to autoregressive (AR) LLMs for text generation, with the potential to decode multiple tokens in a single iteration. However, none of the existing open-source…

Sequence parallelism (SP), which divides the sequence dimension of input tensors across multiple computational devices, is becoming key to unlocking the long-context capabilities of generative AI models. This paper investigates the…