Lichang Chen(@LichangChen2):If you take a look at the index from v4, v4.1, v4.2 to v5, you can see our Muse Spark 1.3 consistently perform better than GPT-5.6. Our RL of 1.3 Max mode is still not fully optimized and we will have a more advanced version in the coming weeks. Stay tuned!

Lichang Chen

@LichangChen2

Code RL & Continual Learning of 🥑🍉 @Meta MSL | Previously: GenAI unit @GoogleDeepmind | PhD’25 @umdcs

가입 September 2021

656 팔로잉 중 1.7K 팬

Lichang Chen@LichangChen2

2026.09.07 19:45

If you take a look at the index from v4, v4.1, v4.2 to v5, you can see our Muse Spark 1.3 consistently perform better than GPT-5.6. Our RL of 1.3 Max mode is still not fully optimized and we will have a more advanced version in the coming weeks. Stay tuned!

248

커뮤니티로 전달

인기 있는 사용자