2026年9月10日

India Open: After a grueling three-game battle, Fa–Min fell to the world No.1 “Phoenix” pair and exited in the quarterfinals.

(New Delhi, Jan 16) The India Open (Super 750) mixed doubles quarterfinals featured Malaysia’s world...

(New Delhi, Jan 16) The India Open (Super 750) mixed doubles quarterfinals featured Malaysia’s world No. 10 pair Goh Soon Huat / Shevon Lai taking on China’s world No. 1 “Phoenix” duo Feng Yan Zhe / Huang Dong Ping. Having lost all four previous meetings, the Malaysians pushed their opponents to three games but eventually fell 19–21, 21–19, 18–21, missing out on a semifinal berth.

In the opening game, the Phoenix pair surged to a 6–1 start. Trailing 2–7, Goh/Shevon responded with a five-point run to close to 7–8, but China still led 11–9 at the interval. After the break, the Malaysians briefly moved ahead 12–11, only for the Phoenix duo to reel off six straight points to stretch to 17–12. Goh/Shevon fought back from 13–19 to 18–19, but China held their nerve to take the first game 21–19.

China again started strongly in the second game, leading 3–0 and then extending to 8–1. Malaysia recovered with four straight points to make it 5–8. At 10–9, China produced a five-point burst to open up 16–9, but the Malaysians responded dramatically with an 8–0 run to turn it around at 17–16. With the score tied at 18–18, Goh/Shevon secured the crucial points to win 21–19 and force a decider.

In the deciding game, the score was level at 2–2 before China edged ahead 4–2 and then built a 9–3 lead. Malaysia rallied to 7–9, but China went into the interval up 11–7. Goh/Shevon later drew level at 13–13 and the game stayed tight to 17–17. In the closing moments, the Phoenix pair won three straight points to reach match point and sealed the match 21–18.

接著讀

ByteDance Unveils GR-3: A General-Purpose Robot “Brain” That Learns New Tasks with Minimal Data and Masters Flexible Objects

On July 22, ByteDance’s Seed team introduced GR-3, a next-generation Vision-Language-Action (VLA) model. Unlike previous systems that require massive robot trajectory datasets, GR-3 can be efficiently fine-tuned on new tasks or unseen objects with just a handful of human demonstrations. It understands abstract language commands and performs dexterous manipulation of soft objects.

415 天前