Trending

Any insider takes on Yann LeCun's push against current architectures?

Block Diffusion: Interpolating Between Autoregressive and Diffusion Models

Owl: Optimized Workforce Learning for multi-agent collaboration

Unraveling Time: A Deep Dive into TTD Instruction Emulation Bugs

Gemma 3 Technical Report [pdf]

A GS-Cache Inference Framework for Large-Scale Gaussian Splatting Models

Experiment with Gemini 2.0 Flash native image generation

Beyond Diffusion: Inductive Moment Matching

Show HN: Nuanced – Help AI understand code structure, not just text

Directly converting skin cells to brain cells yields 1,000% success | Scientists have managed to convert mouse skin cells directly into motor neurons, skipping the usual step of stem cells in between

Gemma 3 Fine-tuning now in Unsloth - 1.6x faster with 60% less VRAM

[R] Multi-View Video Generation via View-Invariant Motion Learning and Cross-View Consistent Translation

Under Trump, AI Scientists Are Told to Remove ‘Ideological Bias’ From Powerful Models A directive from the National Institute of Standards and Technology eliminates mention of “AI safety” and “AI fairness.”

[P] Help with Audio Denoising Model (offline)

roboflow

Save item

notebooks

This repository offers a comprehensive collection of tutorials on state-of-the-art computer vision models and techniques. Explore everything from foundational architectures like ResNet to cutting-edge models like YOLO11, RT-DETR, SAM 2, Florence-2, PaliGemma 2, and Qwen2.5VL.

steven2358

Save item

awesome-generative-ai

A curated list of modern Generative Artificial Intelligence projects and services

modelscope

Save item

FunASR

A Fundamental End-to-End Speech Recognition Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Recognition, Voice Activity Detection, Text Post-processing etc.

NVIDIA

Save item

Megatron-LM

Ongoing research training transformer models at scale

BlinkDL

Save item

RWKV-LM

RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

google-deepmind

Save item

mujoco

Multi-Joint dynamics with Contact. A general purpose physics simulator.