🌟 Photo Sharing Tips: How to Stand Out and Win?
1.Highlight Gate Elements: Include Gate logo, app screens, merchandise or event collab products.
2.Keep it Clear: Use bright, focused photos with simple backgrounds. Show Gate moments in daily life, travel, sports, etc.
3.Add Creative Flair: Creative shots, vlogs, hand-drawn art, or DIY works will stand out! Try a special [You and Gate] pose.
4.Share Your Story: Sincere captions about your memories, growth, or wishes with Gate add an extra touch and impress the judges.
5.Share on Multiple Platforms: Posting on Twitter (X) boosts your exposure an
OPPO collaborates with MediaTek to deploy the AndesGPT edge large model, leading the new era of AI.
AI Technology Express
OPPO has launched the new Xiaobu 1.0 Beta version based on the AndesGPT large model. Users can wake it up and experience various intelligent scenarios by saying "Xiaobu large model."
Adobe unveiled multiple generative AI innovations at the MAX 2023 conference. This includes three new Firefly models that can generate content suitable for commercial use. Adobe also launched over 100 AI feature updates in its flagship Creative Cloud applications, covering software like Illustrator and Photoshop. Adobe Express has added AI features such as generative fill and text-to-template.
Google AI has launched a multi-attribute video dataset named SANPO, suitable for training large models. This dataset consists of real-world and synthetic data, supporting various dense prediction tasks.
Aobi Zhongguang has established a new electronic technology subsidiary in Shunde, Guangdong, with a registered capital of 100 million yuan. The company's business scope includes AI basic software development, application software development, and more.
A study shows that the ChatGPT fine-tuned with synthetic data outperforms GPT-4 in news summarization tasks. Compared to zero-shot GPT-4, the fine-tuned ChatGPT has a win rate of 96%, speed increased by 11 times, and costs reduced by 63%.
Opera browser upgrades AI tool Aria, adding features such as "Refiner" and "Highlighter" to facilitate user interaction with AI, optimize queries, and create content.
The voice synthesis startup ElevenLabs has launched the AI Dubbing feature, which can translate voice into over 20 languages while preserving the speaker's original vocal characteristics.
MediaTek collaborates with OPPO to build a lightweight large model deployment solution on the edge. OPPO's AndesGPT achieves the implementation of large models on the edge through 4-bit quantization technology.
TikTok's new feature supports direct content publishing to the platform from applications like Adobe Premiere Pro, Adobe Express, Twitch, CapCut, and more.
AI startup TabbyML has raised $3.2 million in seed funding, which will be used to develop an open-source code generator.
AI startup Cleanlab has completed a $25 million Series A funding round, achieving a valuation of $100 million. Databricks utilizes Cleanlab to fine-tune the OpenAI model, reducing the error rate by 37%.