Latest News

A focused archive for Latest News. Browse related posts, practical notes, and connected ideas in one stream.

Published posts

Home

Knowledge Collection

Practical Insights

Need ClaudeCode or GPT recharge help?

Use the inquiry pages for ClaudeCode, GPT recharge, assisted purchase, and team sourcing.

Automatic payment is not configured. Pricing, region support, and timing are confirmed manually.

ClaudeCode inquiry GPT inquiry

PortraitGen: Efficient and Diverse Open-Source Portrait Video Editing Tool

PortraitGen is a high-fidelity open-source portrait video editing tool that supports multi-parameter control and 100 FPS rendering. It is suitable for video creation, virtual character design, and similar applications, fulfilling the need for efficient, highly realistic, and personalized creative workflows.

Digital Human

Research

A New Era of Rapid AI Application Development: Exploring Vercel v0, MLE-Agent, and Command R+

The Vercel v0 platform enables developers to build 3D games, interactive applications, and more using natural language in just a few minutes. It supports automatic deployment and hosting, boosting development and sharing efficiency. MLE-Agent serves as an AI engineering intelligent assistant, ideal for managing complex tasks; Command R+ provides RAG optimization and automates multi‑step workflows. By combining v0, MLE-Agent, and Command R+, developers can more efficiently construct, optimize, and manage a diverse range of AI applications.

Large Models

Automation

PaperQA2: Ushering in a Superhuman Era of Scientific Literature Retrieval

PaperQA2 is an open‑source AI tool for scientific literature retrieval that surpasses human experts, developed by Future House. It supports multi‑task processing, including literature search, information extraction, and citation‑network analysis. Evaluated on the LitQA2 benchmark, PaperQA2 delivers outstanding performance in scientific literature retrieval, outperforming researchers at the PhD and post‑doctoral levels. Additionally, the WikiCrow module built on PaperQA2 can generate scientific summaries with accuracy exceeding that of Wikipedia, while the ContraCrow module analyzes contradictions in the literature to help formulate new hypotheses. PaperQA2 pioneers a new mode of interaction with scientific literature, offering researchers an efficient tool for literature analysis.

Large Models

Research

Deepgram Launches AI Voice Agent API: The Future of Real-Time Conversation

Deepgram's newly released AI Voice Agent API delivers seamless real-time voice conversations. Leveraging advanced speech recognition and generation models, the API supports real-time dialogue, pause and interruption handling, and flexible integration with various large language models. Its low latency and strong privacy safeguards make it suitable for scenarios such as customer support and medical transcription.

Large Models

Digital Humans

ElevenLabs Launches the New AI Voice Generation Tool Voice Design: Create Personalized Voices with Text Prompts

ElevenLabs has introduced Voice Design, a cutting‑edge AI voice generation tool that lets users craft personalized speech simply by providing text prompts. Users can tailor attributes such as age, accent, gender, and intonation, and even design voices with mythic or sci‑fi character traits. The solution is ideal for advertising, gaming, podcasts, and more. Voice Design includes fine‑tuning capabilities, integrates seamlessly with ElevenLabs' text‑to‑speech platform, and will soon offer API access and real‑time voice synthesis.

Large Models

Digital Humans

Easily Transform Your Living Room into a VR Scene with VistaDream

VistaDream is an innovative 3D scene generation tool that leverages multi-view consistency sampling technology to create high-quality indoor or outdoor VR scenes from a single photo—without the need for large datasets or complex training. Ideal for VR experiences, interior design, and architectural showcases, it offers a convenient solution for generating immersive scenes.

Research

“Effortless” Is a Myth: Federer’s Philosophy of Success

At the 2024 Dartmouth College commencement, Roger Federer shared his candid perspective on the myth of “effortless”: “Effortless is only a myth. It means training harder, more sweat, beyond self day by day.” He emphasized that true ease stems from daily high‑intensity training and self‑transcendence, and urged graduates to believe in themselves and push their limits. Federer noted that success depends not only on talent but also on perseverance, confidence, and discipline—qualities that are essential to achieving excellence.

Deep Reflection

Adobe Launches Project Turntable: Enables 3D Rotation of 2D Vector Images and Automatically Fills Missing Parts

At Adobe MAX 2024, Adobe unveiled the innovative AI tool Project Turntable, allowing users to rotate 2D vector images as if they were 3D objects, while using generative AI to fill in missing image areas exposed during rotation. The tool streamlines the previously manual process of redrawing perspectives, enabling designers to achieve 3D rotation effects simply by dragging a slider, maintaining image consistency and saving significant time. Project Turntable is currently in early development and is expected to become a standard feature in design tools such as Adobe Illustrator.

Large Model

Adobe's Long-LRM3D and Mamba Architecture: Breakthrough 3D Scene Reconstruction Technology

Adobe's Long-LRM3D leverages the Mamba architecture to reconstruct large-scale 3D scenes from 32 images in just 1.3 seconds. The Mamba architecture integrates mEMBEM and Transformer modules, enabling efficient token processing, merging, and Gaussian pruning, achieving a balance between reconstruction speed and quality. This technology is suitable for large-scale scene reconstruction in gaming, film, and other domains, delivering realistic and efficient visual performance.

Research

Terence Tao on AI: The Future of Mathematics and Collaboration with Artificial Intelligence

Renowned mathematician Terence Tao discusses the potential impact of AI on mathematical research in an interview. He likens current AI tools to a "mediocre but not completely incompetent graduate student," believing AI can accelerate the "industrial-scale" development of mathematics, especially in large‑scale computation and verification. He emphasizes that AI will complement, not replace, mathematicians—handling tedious steps so humans can focus on creative work, particularly in frontier areas.

Digital Marketing

Recommendation

Tesla We Robot Launch Event: Vision for Autonomous Driving and the Future Robot Ecosystem

At the Tesla We Robot launch in October 2024, the company showcased several innovative products, including the brand‑new CyberCab, the Optimus robot, and the Robovan autonomous bus. The CyberCab is a fully driverless two‑seat taxi designed without a steering wheel or pedals, aiming to reduce operating costs to $0.20 per mile. The Optimus robot features high flexibility and AI intelligence, capable of handling heavy physical labor or everyday household chores, with the goal of becoming a household essential. The Robovan, a high‑capacity autonomous transport vehicle, is intended to address high‑density traffic challenges in cities.

Research

Recommendation

PMRF (Posterior-Mean Rectified Flow) is a novel image restoration algorithm that focuses on balancing distortion and perceptual quality in image restoration. It optimizes image quality and reduces distortion through posterior mean prediction and a rectified flow model. Applications of PMRF include denoising, super-resolution, inpainting, and color restoration. Experiments show that PMRF performs excellently on metrics such as PSNR, SSIM, and FID, capable of generating natural, realistic, high‑quality images, representing a significant breakthrough in the current field of image restoration.

Tutorial

1 2 3