agent轨迹让30B打平235B (opens in new tab)

30B agent靠trajectory追平7倍参数模型:ACC把agent解题留下的tool use trajectory重构成长上下文QA对,Qwen3-30B训完MRCR从50.2拉到68.3,跟参数量约7倍的Qwen3-235B-A22B打平 video world model把长期记忆从attention负担里搬出来:WorldKV用retrieval加compression解耦「回头看一眼」的一致性需求,吞吐翻倍且无需fine-tune DiT高分辨率推理转向content-aware scaling:SEGA按spectral energy差异化处理RoPE不同频率成分,避开均匀缩放在结构与细节之间的二选一 8万条terminal录屏反向工程出1530个agent评测任务:TerminalWorld和Terminal-Bench这类专家curate集Pearson仅0.20,过去刷专家集得到的分数可能对不上真实开发者场景

Read the original article
Sign in to keep reading the full article.

Keyboard Shortcuts

Navigation

Next / previous post
j/k
Open post
oorEnter
Preview post
v

Post Actions

Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Discover
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help