微軟公布Orca 2小模型，推論能力不輸大型LLM
· 2023-11-23
參數僅130億的Orca 2模型，可經由模仿更大語言模型的推理方法，在測試中展現不輸給大它10倍模型的推理能力
／微軟
微軟研究人員指出，Orca 2的成功關鍵在於使用多種推論策略，找出不同任務最適用的解決方案，以及使用高品質的合成資料作為後續訓練（post-training）。雖然Orca 2承襲基礎模型及其他模型共通的限制，不過Orca 2突顯了小模型也能具備良好的推理能力，以及可控、可專化，也更具安全性。
热门文章
UK MPs reopen 2025 gambling inquiry as reform stalls
Regulation
GAT CDMX 2025 Institutional Academy: Leaders and Experts Analyze the Present and Future of the Gaming Industry in Mexico and Lat
Sports Game
New Jersey July Gambling Revenue Hits $606M, Sweeps Casinos Banned
Regulation
SBC Summit Canada to Make Player Safety a Key Pillar of 2026 Agenda
Marketing
1spin4win grows its Latin American presence by partnering with Fortuna Juegos
Online Game
1spin4win releases unique slot Don Catleone Hold and Win featuring gangster cats
Online Game
Vietnam’s Controlled Gaming Shift Gains Ground, But Domestic Demand Still Lags
Southeast Asia
Online gambling, crypto pose ongoing money laundering risks in Philippines, analyst says
Southeast Asia
GGC Awards 2026 Shines in Colombo: Honoring Leaders and Innovators in the iGaming Industry
HUIDU Focus
B2B Tech Infrastructure Gains Momentum in Philippine Gaming Sector
Southeast Asia
Kazakhstan plans to penalise online casino promotions
Regulation
HUIDU Invites You to Booth T70 at iGB L!VE 2026 — Let’s Ignite London This July!
HUIDU Focus
Vietnam's tightening online gaming policy creates new market opportunities
Southeast Asia
Institutional Academy that exceeded expectations marked the opening of GAT CDMX
Online Game
GAT Expo Puerto Rico Will Pulse with the New Era of Gaming in the Caribbean
Marketing
