#welm Tech News.
1 story in the last 7 days
The latest welm news, distilled by AI into sharp ~100-word summaries. ByteBrief tracks welm across dozens of tech sources and brings you only what matters, updated hourly. Tap any story for the full brief, or open the original source.
AITechNodeabout 2 hours ago
WeChat AI Scales WeLM to 617B ParametersTencent's WeChat AI team detailed WeLM-HD4-80B and WeLM-HD4-617B models using Hidden Decoding, which expands tokens into internal computation streams without enlarging the Transformer backbone. The 617B model activates 23 billion parameters and outperformed autoregressive baselines across nine benchmarks, with training costs at 4.4 times baseline.
Summaries by ByteBrief