@liljamesjohn @marclou Research shows that AI performance will decline after being blamed.😂
English
Victor Zhang
52 posts
Victor Zhang
@VictorZhangHC
Building @airlingo_ai. Solopreneur at AIR1ST. Ex-Kuaishou • Ex-Tencent • Ex-Anker • Ex-Autel
Katılım Kasım 2024
67 Takip Edilen13 Takipçiler
Testing @airlingo_ai on this funny corporate spin.
Anthropic@AnthropicAI
We've published a paper that explains our views on AI competition between the US and China. The US and democratic allies hold the lead in frontier AI today. Read more on what it’ll take to keep that lead: anthropic.com/research/2028-…
English
@berryxia 这年代benchmark看看就得了，不能真信。我试了几张图，实际效果达不到SOTA，没版式识别，分不清哪里是表格，哪里是图案，两个区域的文字串到一起，还有点幻觉，中文手写比较差。
中文
兄弟们，我后背没有发凉。
但，我看到这个模型架构后高兴不已！
大家还在疯狂堆参数、卷通用大模型的时候，Interfaze直接用一个全新混合架构。
把OCR、视觉、STT、结构化输出这些确定性任务的准确率干到了吊打Gemini-3-Flash、Claude-Sonnet-4.6、GPT-5.4-Mini和Grok-4.3的程度。
他们把任务专用的DNN/CNN编码器和全能transformer融合在一起，实现了真正的“专精+灵活”：CNN负责极致准确和元数据（边界框、置信度）。
transformer负责理解和推理，还能通过标签只激活部分模型，速度和成本直接起飞。
9个硬核基准上全面领先，而且在OCR、视觉、音频这些高频场景里，速度和性价比完胜通用大模型。
这才是最被低估的真相：未来很多真实生产力任务，根本不需要越来越大的通用模型，而是需要这种“专为确定性任务而生”的混合架构。
Interfaze把AI从“万能但贵且慢”拉回了“又准又快又便宜”的现实路径。
PS： 我觉得需要和我之前的ocr测试的用例看看一起再测试一下结果怎么样，是不是如榜单所描述！
完整博客值得立刻读一遍👉 interfaze.ai/blog/interfaze…
Interfaze (YC P26)@interfaze_ai
Interfaze is a new model architecture that outperforms in tasks like OCR, Object detection, Translation and more It beats models like Claude Sonnet 4.6, Gemini 3 flash and GPT 5.4 mini on 9 benchmarks
中文
@bamboobee5 @yetone 感谢关注，Airlingo是帮助第二语言深度阅读和学习的Chrome插件，在开发的最后阶段了。
中文
@PandaTalk8 张小珺的采访功力真不大行，听得最多的问题是八卦式打探，很少让受访者和观众惊喜的好问题，比起记者Bob Cringely在The Lost Interview问Steve “How do you know what’s the right direction?”，真是差了十万八千里。
中文
张小珺的播客成了中国顶尖AI科学家们展示自己的最大展台了。
张小珺 Xiaojun Zhang@zhang_benita
This episode features an interview with Yao Shunyu @ShunyuYao14 , Research Scientist at Google DeepMind. Yao has held research scientist roles at both Anthropic and Google DeepMind, contributing to the development of key models including Claude 3.7, 4.5, and Gemini 3. Yao Shunyu is not your typical nerd. Every now and then, he’ll catch you off guard with a flash of irreverence. “None of the old guard are your relatives — so if you think someone’s being dumb, they’re just being dumb. Say it. No big deal.” (laughs) “Everyone’s a surfer now, but what really matters is the wave — not the person riding it.” “AI doesn’t actually require that much brainpower — I mean it genuinely doesn’t — most of this is work any undergrad could do. The most important quality in this industry is reliability: being meticulous, and taking responsibility for what you put out.” “You don’t need to worry too much about ruffling feathers with your opinions. As long as your views are internally consistent — not just taking random shots at people, but grounded in your own genuine understanding — there are objective standards for how you’re doing in this field. People will respect you for it.” Let us have a little fun with this one! 😄 youtu.be/ttkd0t5qTD4?si…
中文
@jeetnirnejak @paper @motiondotdev Great design. I made a similar component to show the user the payment methods they've saved. It's an animated carousel for the card stack, vibe coded with codex and @motiondotdev
English
A wallet component, spent too long on getting the spring physics right
Designed in @paper, animated with @motiondotdev
English
@jakubkrehel I've added a badge to show the card state and a flip card animation to show the billing address.
English
@Stanleysobest 这比赛确实很刺激，赛道高速后的急弯多，61号有个急弯失误，75号顺势想超他，结果缠斗在一起速度都没拉到最大，Debis说那个超车的弯是全油门走另外的线。不得不说姜还是老的辣，战场瞬息万变，Debis能迅速抓住机会。
中文
@SaitoWu 因为要巩固市场地位，才能躺着赚支付抽成，像他们的billing系统历史包袱太重，实在是改不动，跟不上usage billing的浪潮，所以直接收购了Metronome
中文
@thsottiaux Thank you for resetting it. I'm using GPT-5.4 for an ongoing task, but the Codex extension in VS Code keeps turning on fast mode automatically, even after I've turned it off multiple times. Is this a bug, or is it designed to encourage more credit usage?
English
Don't just reset Codex rate limits for fun, it costs money.
Don't just reset Codex rate limits for fun, it costs money.
... but the vibes are good ...
I have reset Codex rate limits for ALL paid plans to celebrate a good week and allow everyone to build more with GPT-5.5. Enjoy
English
Using the TTS model with character-based pricing for Chinese speech is much more cost-effective than for English.
Aran Komatsuzaki@arankomatsuzaki
Follow-up on non-English token-inefficiency with more model-language pairs: - Chinese is cheaper than English on major Chinese models - Gemini and Qwen provide least non-English tax - Anthropic has the highest tax by far; Kimi is next - Hindi is the worst-covered language here, despite its massive speaker base
English
发现中文推关于Manus事件的众多内容和媒体圈高度相似。同样一个大新闻，要么抢时效转发，要么加一些评论，反复翻炒博取流量，不同的是多了媒体圈没有的臆测，但少了媒体圈才有的深度内容。
中文
綜合來看，平均表現最好的模型是 Google 專門針對翻譯任務微調的 translategemma:27b；單就繁體中文翻譯來說則是 qwen3.5:35b。 #note
微調 TranslateGemma 官方其實有釋出技術報告，把 Gemini 大模型的「翻譯直覺」蒸餾進開放架構的小模型。
Translation Quality Benchmark
github.com/hydropix/Trans…
中文
