AI Agent Skills

[NeurIPS 2025] Flow x RL. "ReinFlow: Fine-tuning Flow Policy with Online Reinforcement Learning". Support VLAs e.g., Pi0, Pi0.5, GR00TN1.5. Fully open-sourced.

UFM: A Unified Dense Image Correspondence Estimator for both Optical Flow & Wide Baseline Matching Tasks. Matches any pair of images. (NeurIPS 2025)

[CVPR 2024 & NeurIPS 2024] EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI

[NAACL 2021] QAGNN: Question Answering using Language Models and Knowledge Graphs 🤖

🔥[NeurIPS'25] DeepFund: Pilot for Your Next Fund Investment

[NeurIPS 2024] PointMamba: A Simple State Space Model for Point Cloud Analysis

A PyTorch Library for Photonic AI Computing Model Training and Co-Design (NeurIPS'21)

[NeurIPS 2024] ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution

[NeurIPS 2025] VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models

[NeurIPS 2025 & ICLR 2025 Financial AI Best Paper Award] A multi-agent framework that leverages LLMs to simulate socio-economic systems

Official repository for "RLVR-World: Training World Models with Reinforcement Learning" (NeurIPS 2025), https://arxiv.org/abs/2505.13934

[NeurIPS 2025] More Than Generation: Unifying Generation and Depth Estimation via Text-to-Image Diffusion Models

[NeurIPS 2024 D&B] GTA: A Benchmark for General Tool Agents & [arXiv 2026] GTA-2

Official repository for "iVideoGPT: Interactive VideoGPTs are Scalable World Models" (NeurIPS 2024), https://arxiv.org/abs/2405.15223

A ComfyUI node pack that implements FreeLong (NeurIPS 2024) spectral blending for Wan 2.2 video generation

AlgoTune is a NeurIPS 2025 benchmark made up of 154 math, physics, and computer science problems. The goal is write code that solves each problem, and is faster than exi…