AI Train2026
End-to-end Python framework for training and serving NVIDIA's Cosmos world models (Cosmos3), integrating distributed training (FSDP/TP/CP/PP), DCP/safetensors checkpoints, dataset adapters, multiple inference backends, online serving, and agent skills.
AI Train2026
Compresses an entire document collection into 16 query-dependent memory tokens for a frozen LLM. It retrieves useful clues across long histories while keeping the model’s input length fixed.
Studies train-time knowledge injection via hypernetworks that generate fixed LoRA adapters from large fact corpora, empirically characterizing power-law scaling across hypernetwork depth, width, and target model size and reporting improved OOD generalization.
AI Dataset2026
Turns chemistry and molecular-property questions into verifiable short-answer tasks for language-model training and evaluation, with nearly 50,000 questions drawn from multiple datasets.
AI Video2026
Enables native joint video-and-audio generation and training up to 2K using a dynamic block-sparse attention scheme that cuts attention cost while preserving audio–visual coupling; provides preview checkpoints, training and inference code. High-resolution training requires large multi-GPU (80GB) clusters and CUDA ≥ 12.4.
AI Agent Papers2026
Chengyi Yang, Renxu Bao +11Affiliation: Xiamen University, Affiliation: Washington University in St. Louis +2
Converts real repositories into verifiable, sandboxed training environments with black-box tests to train coding agents on long-horizon program reproduction. Provides Program-Env for automated environment construction and Program-Agent for a role-isolated development workflow, yielding 1,648 C/C++/Go/Rust tasks and improving benchmark performance.
AI Dataset2026
Provides synthetic task environments for training data-analysis agents, with self-contained tasks and evaluation rubrics grounded in known generative mechanisms.
AI Dataset2026
Provides self-contained synthetic environments for training data-analysis agents. This slice contains 1,538 tasks, each graded with a rubric designed to evaluate analytical findings.
AI Model2026
Fine-tunes Qwen3.8-27B to change how it answers politically sensitive questions, targeting both factual recall and framing. The training reinforces values selected by Nous Research; intended for users who want that viewpoint or study value alignment.
AI Train2017
Trains vision, text, tabular, recommendation, and medical imaging models through a layered PyTorch API. High-level learners cover common workflows, while lower layers stay available when researchers need custom behavior.