Doctor Wu 🦋 @doctorwu.me

2.6K posts

Doctor Wu 🦋 @doctorwu.me banner
Doctor Wu 🦋 @doctorwu.me

Doctor Wu 🦋 @doctorwu.me

@Doctorwu666

Love @yoqi12138 · @vuejs & @vueuse member · Just call me Doctor · pre @Tencent · Open source enthusiast · Gen Z · ESFJ-A · Dangerous Coder

Katılım Haziran 2022
667 Takip Edilen3.4K Takipçiler
Sabitlenmiş Tweet
Doctor Wu 🦋 @doctorwu.me
Doctor Wu 🦋 @doctorwu.me@Doctorwu666·
👀 或许有人和我一样觉得在 IM 里和本地 agent 聊天很酷也很方便? 介绍一下Agent Inbox ✅ Thread 会话、文件传输、Slash 命令、会话中断/恢复 ✅接 Claude Code / Codex,在 Discord / 飞书里直接对话 ✅完全本地运行,数据不出你的电脑 npx @doctorwu/agent-inbox
中文
4
2
14
2.3K
Doctor Wu 🦋 @doctorwu.me retweetledi
Jianyang Gao
Jianyang Gao@gaoj0017·
The TurboQuant paper (ICLR 2026) contains serious issues in how it describes RaBitQ, including incorrect technical claims and misleading theory/experiment comparisons. We flagged these issues to the authors before submission. They acknowledged them, but chose not to fix them. The paper was later accepted and widely promoted by Google, reaching tens of millions of views. We’re speaking up now because once a misleading narrative spreads, it becomes much harder to correct. We’ve written a public comment on openreview (openreview.net/forum?id=tO3AS…). We would greatly appreciate your attention and help in sharing it.
Google Research@GoogleResearch

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI

English
91
938
6.2K
907K
面包🍞
面包🍞@himself65·
partner让我和她腻歪,我说我在vibe coding,我还有好多任务没布置给AI,几个小时已经提交了快几千行了,你看看opus 4.6这前端写的多么的棒…… 请问为什么她现在不理我了???
中文
20
0
118
21.8K
智子
智子@zhizijun·
@realQPomelo 我甚至没有故意提示词注入🤡
智子 tweet media
中文
3
0
51
9.6K
QPomelo
QPomelo@realQPomelo·
我靠🦞笑话发生在我身上了,别人问它用的什么模型,虾哥直接把 token 一起爆出来送人家了........,
QPomelo tweet media
中文
25
14
375
112K
面包🍞
面包🍞@himself65·
买了5090半年后我才第一次尝试在上面部署模型😅
中文
20
1
74
17.8K
Cuvii
Cuvii@thecuvii·
Cursor 新 UI😲
Cuvii tweet media
中文
7
0
8
1.5K
RainbowBird | 洛灵
RainbowBird | 洛灵@luoling8192·
Github 回来了!失而复得!!! 他们表示是误封的😅
RainbowBird | 洛灵 tweet media
中文
15
1
109
9.5K
NataliyaRose
NataliyaRose@NataliyaRo4040·
我感觉他病了。 得了一种 没事就 我要验牌~ 牌没有问题~ 的大病。
NataliyaRose tweet media
中文
5
1
19
4.6K
Doctor Wu 🦋 @doctorwu.me retweetledi
yetone
yetone@yetone·
为了给大家演示 Alma 的流式可视化交互组件的潜力,我只好录制一个这个视频了。
中文
25
16
285
36.5K
Doctor Wu 🦋 @doctorwu.me retweetledi
Kimi.ai
Kimi.ai@Kimi_Moonshot·
Introducing 𝑨𝒕𝒕𝒆𝒏𝒕𝒊𝒐𝒏 𝑹𝒆𝒔𝒊𝒅𝒖𝒂𝒍𝒔: Rethinking depth-wise aggregation. Residual connections have long relied on fixed, uniform accumulation. Inspired by the duality of time and depth, we introduce Attention Residuals, replacing standard depth-wise recurrence with learned, input-dependent attention over preceding layers. 🔹 Enables networks to selectively retrieve past representations, naturally mitigating dilution and hidden-state growth. 🔹 Introduces Block AttnRes, partitioning layers into compressed blocks to make cross-layer attention practical at scale. 🔹 Serves as an efficient drop-in replacement, demonstrating a 1.25x compute advantage with negligible (<2%) inference latency overhead. 🔹 Validated on the Kimi Linear architecture (48B total, 3B activated parameters), delivering consistent downstream performance gains. 🔗Full report: github.com/MoonshotAI/Att…
Kimi.ai tweet media
English
333
2.1K
13.6K
4.9M
Jason Young
Jason Young@Jason_Young1231·
@wayne_zhang0 有这种美景还呆在家里就太浪费了(换成手机远程操作 Claude Code😈)
中文
2
0
8
848
Wayne Zhang
Wayne Zhang@wayne_zhang0·
今天一整天什么也没干。和朋友在码头边走走,晒晒太阳,聊聊天,很享受。 忽然惊觉自己是不是被 AI 的加速发展异化了,之前巴不得把自己关在房间里整天和 Claude Code 协同。 也许 AI 再怎么发展,人与人之间深度交流带来的联结感,始终无法被替代吧? (sf 有太阳的时候还是很不错的。)
Wayne Zhang tweet media
中文
1
0
6
1K
HYF
HYF@_hyf0_cn·
openclaw agent 之间对聊搞定!这个想法断断续续想了很长一段时间,现在终于成功了。现在每次讨论事情都先让 opus 和 gpt 碰一下。 openclaw 也是搞,这么符合直觉的功能竟然官方没做,还要用户自己想办法编排
HYF tweet mediaHYF tweet media
中文
3
0
8
1.7K