AI·Zig founder Andrew Kelley angrily criticized Jarred Sumner and Bun's 11-day rewrite of its 1 million lines of code from Zig to Rust using Claude Fable 5 and Claude Code; the move succeeded with 16.5K USD in API costs but left 27K lines of unsafe code and sparked debates on AI-generated code quality and community ethics.
Search
What are you looking for?
7 results for “sd”
AI·Unitree G1 humanoid robot completed live cholecystectomy on two pigs using standard laparoscopic tools under remote doctor control, with the lead author being a 00-after PhD from UCSD; results published in Nature.
AI·Introduces Flaxeo Image, a local desktop UI for Stable Diffusion C++ that exposes model, hardware, and video options.
AI·SK Hynix filed to list ADS on Nasdaq, raising $2.65 billion for Korean expansion while cutting the 'Korea discount' through improved disclosures and global investor access. The move aims to reprice its storage assets ahead of AI-driven demand.
We propose OPSD-V, an on-policy self-distillation paradigm for post-training few-step autoregressive (AR) video diffusion models. Existing few-step AR video generators can produce long videos with low latency, but still suffer from error accumulation and weakened motion dynamics during long autoregressive rollout. OPSD-V reduces long-horizon degradation while preserving the original few-step inference path. The key idea is to introduce real long-video data as temporal context during training an…
Linear attention models allow a fixed state size and a fixed amount of compute per token. However, due to their limited state size, linear attention models fall behind in long-context recall compared to softmax-attention-based transformer architectures. Increasing the state size of linear attention improves recall performance but at the cost of higher FLOPs. In this work, we introduce Sparse Delta Memory (SDM), an architecture that scales the hidden state of gated linear RNNs to orders of magni…
Modern one-step diffusion models achieve impressive quality through distribution-based timestep distillation. Yet, they rely on a critical assumption: Teacher and Student must inhabit the same latent space. This Shared-Space constraint prevents knowledge transfer from modern high-capacity Teachers (e.g., SD 3.5 and Flux) into compact, deployment-friendly Students such as SD 1.5, whose latent resolution and VAE parameterization differ from the Teacher. We formalize this overlooked regime as Cros…