Some wins are about who can cross first. Today was about how they got there. @BrandonMcNult and @adamyates7 set the pace. @TamauPogi had the line. @ISAACDELTOROx1 got the moment: his first Tour stage win and a @TeamEmiratesUAE one-two.

TradFi x Crypto Finance took center stage at #Web3Festival#. 🇭🇰 Our CEO @mehler joined industry leaders across finance and crypto to discuss the future of cross-border payments and asset digitization, sharing how @Stable is building payment rails for faster global settlement. 📢 Fellow panelists: Richard Teng @_RichardTeng, Co-CEO of @binance Akhil Devmurari, Head of Fintech Industry for Payments, @jpmorgan Asia Pacific Leonard Hoh @Len_hoh, President of @Bitstamp Global payments are entering a new era. 🌍

Through institutionalized cross-regional collaboration, China channels teaching staff, management expertise, and educational resources from more developed eastern regions to western ethnic regions. This approach is known as educational "paired assistance" and is positioned as a government-led model for promoting educational equity. #DeepChina# Taking #Xizang# and #Xinjiang# as examples, in 2016, the "group-style" educational talent assistance program for Xizang was officially launched. The "group-style" model involves sending teams of 10 to 50 teachers from 17 partner provinces, as well as affiliated primary and secondary schools of universities directly under the Ministry of Education of China, to provide concentrated support to individual primary or secondary schools in Xizang. The first cohort comprised more than 800 teachers. Over the past decade, more than 2,500 dispatched teachers have provided paired support to 21 primary and secondary schools in Xizang. Dispatched teachers have formed nearly 3,000 mentoring pairs with local teachers, assisted in more than 200 teaching teams. Meanwhile, the government facilitated the participation of more than 2,700 local teachers in on-the-job training in partner provinces and cities, building a core teaching force that "will not leave." #EthnicUnity# In 2021, the "group-style" educational talent selection program for #Xinjiang# was launched, fully leveraging the quality educational resources of partner provinces and cities to send outstanding educational professionals to Xinjiang to strengthen its teaching force. One "group-style" team consists of approximately 10 full-time teachers and educational administrators, providing concentrated support to one school. The recipient level covers the four educational stages of kindergarten, primary school, middle school, and high school (including secondary vocational schools). #ChineseNation#

The hip-hop performance Joy of the Present Moment takes the stage. Young dancers from Xizang spread joy with dynamic moves, strengthening cross-cultural communication through trendy street dance #HipHopPerformanc#

If Dalio is right and we’re entering a Stage 6 world of capital wars, asset freezes and geopolitical fractures… What’s the cleanest vehicle to preserve mobility of capital? Something that: - Can’t be seized by a foreign government - Can’t be frozen by a bank - Can cross borders instantly - Doesn’t depend on any single state There’s really only one answer.

Late Bear Market Signals Are Flashing 🐻 Multiple indicators suggest Bitcoin is entering the late stages of a bear market, but it is still too early to call a bottom. I want to see the CryptoQuant Bull-Bear Market Cycle Indicator 30DMA cross above the 365DMA first.

Step into our Umamusume: Pretty Derby panel, hosting star guest voice actresses Tomoyo Takayanagi, who brings Oguri Cap to life, and Naomi Ohzora, the voice behind Tamamo Cross. Make your way over to the Main Stage on Sunday, 30 August 2026 at 1:30 PM for an engaging stage presentation where our guests will share their experiences, dive into character lore, and reflect on their journey with the title.

The panel on "Re-engineering Global Money Rails: APAC, Interoperability & Scale" just wrapped at @leapeast and what a conversation it was. 🙌 Our Global CEO @50Nent (Nenter Chow) joined an incredible group of minds on the Orbital Stage at HKCEC to dig into the future of #crypto-enabled# cross-border payments, interoperability, and what it takes to build financial infrastructure at scale across APAC and beyond. A huge thank you to our fellow panelists: • Elliott Gotkine (@ElliottGotkine) for moderating with sharp, incisive questions • Nicole Sandler, Chief Ecosystem Officer at @ubyx_ • Dr. Yanan Wu, Chairman & CEO at @SurfinMeta • Colin Payne, Head of Innovation at @TheFCA The conversation was real, the room was packed, and the ideas were bold. This is exactly the kind of dialogue that moves the industry forward. 📍 Hong Kong Convention & Exhibition Centre #LEAPEast#

TL;DR XPeng released X-AuT, a compression method that shrinks a speech LLM's audio encoder from 18 to 14 layers with barely any accuracy loss — and the 16-layer version actually improves accuracy. Title: X-AuT: Progressive Audio-Encoder Compression for Speech LLMs with Cross-Scale Distillation URL: Points ✂️ It uses "progressive pruning," cutting the audio encoder from 18 to 16 to 14 layers in stages 🔍 What's cool: single-layer removal scores alone can't predict the best pair, so it explicitly measures layer-pair interactions 🎓 Cross-scale distillation from a 1.7B teacher into a 0.6B student clearly beats same-scale self-distillation 📈 The 16-layer model cuts parameters by 10.35% while improving macro-average error from 5.61% to 5.27% 📉 The 14-layer model cuts parameters by 20.70% with only a +0.14pt accuracy drop 🚗 On an in-vehicle chip, it cuts encoder time by 21.4% and total latency by 4.7% 🔓 Code and models are already public on GitHub and Hugging Face (CC BY-NC 4.0) What stands out is the careful design of "how to prune smart and recover well," not just cutting layers. #SpeechLLM# #ModelCompression#