Synced 机器之心

New Language Model Breakthrough on Nature: FunSearch Addresses A Longstanding Mathematics Challenge In a recent paper titled "Mathematical Discoveries from Program Search with Large Language Models, a research team introduces FunSearch—a novel approach that elevates LLM-guided evolutionary procedures. FunSearch not only achieves breakthroughs in established open problems but also leads to the discovery of new algorithms. Latest from Synced Review https://lnkd.in/e_Wg882A

Tencent’s FaceStudio Redefines Image Generation with Identity-Preserving Efficiency in Seconds A recent paper from Tencent's research team introduces a novel identity-preserving synthesis approach, with a specific focus on human images. The proposed model adopts a direct feed-forward mechanism, eliminating the need for intensive fine-tuning and streamlining the image generation process. Latest from Synced Review https://lnkd.in/gn46XZSu

Microsoft’s TaskWeaver: Empowering Intelligent Conversational Agents for Handling Domain-Specific Complex Tasks A Microsoft research team introduces TaskWeaver, a cutting-edge, code-first framework designed to empower LLM-powered autonomous agents. TaskWeaver offers a potent and flexible platform for constructing intelligent conversational agents capable of handling complex tasks and seamlessly adapting to domain-specific scenarios. Latest from Synced Review https://lnkd.in/gR2Bg-aP

Tencent & Sydney U’s GPT4Video: A Unified Multimodal Large Language Significantly Elevates LMs’ Video Generative Capabilities A collaborative effort between Tencent AI Lab and The University of Sydney introduces GPT4Video, which stands as a unified multi-model framework that endows Large Language Models (LLMs) with the unique ability for both video understanding and generation. Latest from Synced Review https://lnkd.in/gbZDHyYF

Spatial-Temporal Innovation: STLVQE Redefines Real-Time Video Enhancement for an Unmatched Viewing Experience A paper titled "Online Video Quality Enhancement with Spatial-Temporal Look-up Tables" introduces a novel method, STLVQE. This research, conducted by a team from Tongji University and Microsoft Research Asia, pioneers the exploration of the online video quality enhancement problem and presents the first method achieving real-time processing speed. Latest from Synced Review https://lnkd.in/gBtJ8hHx

Adobe’s DMV3D Achieves SOTA Performance for High-Fidelity 3D Objects Generation Within Seconds A research team innovative single-stage category-agnostic diffusion model. This model can generate 3D Neural Radiance Fields (NeRFs) from either text or a single-image input condition through direct model inference, enabling the creation of diverse high-fidelity 3D objects in just 30s/asset. Latest from Synced Review https://lnkd.in/g7fkhnSn

DeepMind’s DiLoCo Revolutionizes Language Model Training with 500× Less Communication In a new paper DiLoCo: Distributed Low-Communication Training of Language Models, a Google DeepMind research team presents Distributed Low-Communication (DiLoCo). DiLoCo employs a distributed optimization algorithm that facilitates the training of language models on islands of poorly connected devices, surpassing the performance of fully synchronous models while reducing communication by 500 times. Latest from Synced Review https://lnkd.in/etFSt7PV

Meet LEO: An Embodied Generalist Agent Excelling in 3D World Tasks In a new paper An Embodied Generalist Agent in 3D World, a research team introduces LEO, which stands as an embodied multi-modal and multi-task generalist agent that excels in essential capabilities such as perception, grounding, reasoning, planning, and action within the intricate 3D world. Latest from Synced Review https://lnkd.in/g46CYe8y

Meet LEO: An Embodied Generalist Agent Excelling in 3D World Tasks In a new paper An Embodied Generalist Agent in 3D World, a research team introduces LEO, which stands as an embodied multi-modal and multi-task generalist agent that excels in essential capabilities such as perception, grounding, reasoning, planning, and action within the intricate 3D world. Latest from Synced Review https://lnkd.in/g46CYe8y

ETH Zurich’s UltraFastBERT Realizes 78x Speedup for Language Models In a new paper Exponentially Faster Language Modelling, an ETH Zurich research team introduces UltraFastBERT, a variant of the BERT architecture. UltraFastBERT takes a revolutionary approach by replacing feedforward layers with fast feedforward networks, resulting in an impressive 78x speedup over the optimized baseline feedforward implementation. Latest from Synced Review https://lnkd.in/gXv6RD5s