CPGD: Toward Stable Rule-based Reinforcement Learning for Language Models CPGD introduces a novel reinforcement learning algorithm designed to stabilize policy updates for language models trained with rule-based rewards, addressing instability and training collapse issues found
RESEARCH
-

vec2vec: Unsupervised Text Embedding Translation Method
By
–
Harnessing the Universal Geometry of Embeddings This work introduces vec2vec, the first unsupervised method to translate text embeddings between different vector spaces without any paired data, encoders, or predefined matches, by leveraging a universal latent semantic
-

MAGI-1: Large-Scale Autoregressive Video Generation Model
By
–
MAGI-1: Autoregressive Generation at Scale MAGI-1 is a large-scale autoregressive video generation model that produces videos by generating fixed-length chunks of consecutive frames with diffusion, enabling efficient, streaming-capable, and temporally consistent video
-

MMaDA: Multimodal Diffusion Language Models for Reasoning
By
–
MMaDA: Multimodal Large Diffusion Language Models MMaDA introduces a new class of multimodal diffusion models that unify reasoning and generation across text and vision, achieving state-of-the-art performance in both understanding and generation tasks. Problem: Existing
-

Panda: Pretrained Model for Chaotic Dynamics Forecasting
By
–
Panda: A pretrained forecast model for universal representation of chaotic dynamics Panda is a pretrained attention-based model trained solely on synthetic chaotic systems, capable of zero-shot forecasting across diverse, unseen real-world dynamical systems, including PDEs it
-

Robin: Multi-Agent AI System Automates Scientific Discovery Loop
By
–
Robin: A multi-agent system for automating scientific discovery Robin is the first multi-agent AI system to automate the entire scientific discovery loop—literature review, hypothesis generation, experiment design, data analysis, and hypothesis refinement—within a single
-

GUI-G1 and Robin: High-Performance Agents with Multimodal Improvements
By
–
Major upgrades for high-performance agents with the rise of GUI-G1 and Robin, alongside the notable multimodal improvements of diffusion language models Check out the top 10 papers for the week – GUI-G1: Understanding R1-Zero-Like Training for Visual Grounding in GUI
-

II-Researcher: Advanced Research Assistant with LangChain Integration
By
–
II-Researcher A research assistant that combines search providers and web scraping with LangChain's text processing functionality to tackle complex questions. Features flexible LLM support and comprehensive data gathering capabilities. Explore on GitHub!
-
Google positioned to achieve AGI first due to existential threat
By
–
Google will be the first one to get AGI because they are the ONLY big tech company for whom the rise of AI is an imminent existential threat.
-

First Digital Image Scanned in 1957 Launched Raster Graphics
By
–
The first digital image was scanned #otd in 1957. The late Russell Kirsch’s grayscale image of his newborn son launched the field of raster graphics: https://
bit.ly/36tpptN
