{"type": "article", "title": "Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/flash-dllm-io-aware-kv-caching-and-parallel-decoding-for-fast-memory-efficient", "original_source": "https://aiflash.com/news/124713/", "published": "2026-09-23T02:30:25+00:00", "accessed": "2026-09-23", "id": "flash-dllm-io-aware-kv-caching-and-parallel-decoding-for-fast-memory-efficient"}