【AI最新動向 2026年2月14日】論文5件・GitHub5件

2026年2月14日

📝 この記事のポイント

🚀 AI技術の最新動向 - 2026年2月14日世界中から収集したAI・機械学習の最新情報をお届けします 📑 目次💻 注目のGitHubプロジェクトmilanm/AutoGrad-EngineKuberwastaken/picogptScottT2-spec/mnist-neural-network-ScottT2-spec/malaria-cell-detectionobsTR/Sherlock📌 関連記事もチェック📚 最新研究論文1. Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment著者: Jacky Kwok, Xilun Zhang, Mengdi XuThe long-standing vision of general-purpose robots hinges on their ability to understand and act upon natural language instructions. Vision-Language-Action (VLA) models have made remarkable progress toward this goal, yet their generated actions can still misalign with the given instructions. In this...論文を読む →2. UniT: Unified Multimodal Chain-of-Thought Test-time Scaling著者: Leon Liangyu Chen, Haoyu Ma, Zhipeng FanUnified models can handle both multimodal understanding and generation within a single architecture, yet they typically operate in a single pass without iteratively refining their outputs. Many multimodal tasks, especially those involving complex spatial compositions, multiple interacting objects, o...論文を読む →3. AttentionRetriever: Attention Layers are Secretly Long Document Retrievers著者: David Jiahao Fu, Lam Thanh Do, Jiayu LiRetrieval augmented generation (RAG) has been widely adopted to help Large Language Models (LLMs) to process tasks involving long documents. However, existing retrieval models are not designed for long document retrieval and fail to address several key challenges of long document retrieval, includin...論文を読む →💻 注目のGitHubプロジェクト1. milanm/AutoGrad-EngineA complete GPT language model (training and inference) in pure C# with zero dependencies. ⭐ 177 stars | 🔀 21 forksリポジトリを見る →2. Kuberwastaken/picogptGPT in a QR Code ; The actual most atomic way to train and inference a GPT in pure, dependency-free Python.⭐ 40 stars | 🔀 4 forksリポジトリを見る →3. ScottT2-spec/mnist-neural-network-Neural network from scratch (NumPy only, 96% accuracy) + Kaggle Digit Recognizer competition entry (99.685% accuracy, top 40). No frameworks for the from-scratch version.⭐ 14 stars | 🔀 1 forksリポジトリを見る →4. ScottT2-spec/malaria-cell-detectionCNN-based malaria detection from blood cell microscope images — 95.43% test accuracy on NIH dataset (27,558 images)⭐ 10 stars | 🔀 0 forksリポジトリを見る →5. obsTR/SherlockMulti modal deepfake detection system using FastAPI, PyTorch, and a Cloudflare Workers edge gateway, with a Next.js dashboard.⭐ 10 stars | 🔀 0 forksリポジトリを見る →。

🚀 AI技術の最新動向 – 2026年2月14日

世界中から収集したAI・機械学習の最新情報をお届けします

📑 目次

💻 注目のGitHubプロジェクト
1. milanm/AutoGrad-Engine
2. Kuberwastaken/picogpt
3. ScottT2-spec/mnist-neural-network-
4. ScottT2-spec/malaria-cell-detection
5. obsTR/Sherlock
📌 関連記事もチェック

📚 最新研究論文

1. Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment

著者: Jacky Kwok, Xilun Zhang, Mengdi Xu

The long-standing vision of general-purpose robots hinges on their ability to understand and act upon natural language instructions. Vision-Language-Action (VLA) models have made remarkable progress toward this goal, yet their generated actions can still misalign with the given instructions. In this…

論文を読む →

2. UniT: Unified Multimodal Chain-of-Thought Test-time Scaling

著者: Leon Liangyu Chen, Haoyu Ma, Zhipeng Fan

Unified models can handle both multimodal understanding and generation within a single architecture, yet they typically operate in a single pass without iteratively refining their outputs. Many multimodal tasks, especially those involving complex spatial compositions, multiple interacting objects, o…

論文を読む →

3. AttentionRetriever: Attention Layers are Secretly Long Document Retrievers

著者: David Jiahao Fu, Lam Thanh Do, Jiayu Li

Retrieval augmented generation (RAG) has been widely adopted to help Large Language Models (LLMs) to process tasks involving long documents. However, existing retrieval models are not designed for long document retrieval and fail to address several key challenges of long document retrieval, includin…

論文を読む →

💻 注目のGitHubプロジェクト

1. milanm/AutoGrad-Engine

A complete GPT language model (training and inference) in pure C# with zero dependencies.

⭐ 177 stars | 🔀 21 forks

リポジトリを見る →

2. Kuberwastaken/picogpt

GPT in a QR Code ; The actual most atomic way to train and inference a GPT in pure, dependency-free Python.

⭐ 40 stars | 🔀 4 forks

リポジトリを見る →

3. ScottT2-spec/mnist-neural-network-

Neural network from scratch (NumPy only, 96% accuracy) + Kaggle Digit Recognizer competition entry (99.685% accuracy, top 40). No frameworks for the from-scratch version.

⭐ 14 stars | 🔀 1 forks

リポジトリを見る →