Lewis N Watson retweeted
HAI2026本会議はじまりました!昨年とのトレンド比較だと、まず投稿が1.5倍(200件以上!)になっているが、トレンドの違いもあるとのこと。delegationに関する研究が増えているのは興味深い(しかしここ数年の投稿の増加は驚異的!)投稿では従来日本と欧州が多かったが、米国と中国がだいぶ増えている
Lewis N Watson retweeted
We're giving away a Hogwarts Legacy custom wrapped NVIDIA GeForce RTX 5090 to celebrate its official release with DLSS 5 🪄
To enter:
🟢 Share this post
🟢 Comment #RTXON
Lewis N Watson retweeted
We're giving away a Hogwarts Legacy custom wrapped NVIDIA GeForce RTX 5090 to celebrate its official release with DLSS 5 🪄
To enter:
🟢 Share this post
🟢 Comment #RTXON
Terms and Conditions: nvidia.com/en-us/geforce/con…
Lewis N Watson retweeted
Replying to @bunkaich
You're not imagining it - it's a grounding issue. If you unplug from the charger note how it goes away.
It's because the plug on the modern chargers is not connected to the grounding pin. It used to be metal inside the plug but then they switched it to plastic. The charger itself still is capable of being grounded but the plug is preventing it.
I just took a metal wire, wrapped it around the grounding pin and then tucked it inside to touch the ground on the charger - and that solved the entire issue - no more tingling.
Lewis N Watson retweeted
Absolutely incredible: MIT researchers say GPT-6 Astra improved a small Qwen model’s math score from 7.9% to 48.1% by directly editing just 0.016% of its weights, without gradient-based training.
The result points toward AI systems that can diagnose and improve other models directly, a potential step toward self-improving AI.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Lewis N Watson retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Lewis N Watson retweeted
In lieu of flowers, memorial contributions may be made to:
Johns Hopkins Alzheimer’s Disease Research Center Fund
alzresearch.org/giving/
Alzheimer’s Foundation of America Caregiver Resources
alzfdn.org/donate/
Photo Credit: Tommy Garcia
Meet NVIDIA Nemotron 3.5 Transcribe 🎙️
An open-weight, promptable 30B MoE model with ~4B active parameters that brings transcription in 40 languages, speaker attribution, translation, contextual prompting and audio understanding into one checkpoint.
Watch it keep track of who said what when everyone’s talking at once.
1/ 🧵
Lewis N Watson retweeted
its really fucking evil
if you have a qwen3.8 27b around, try steering L34 N3838 +0.75, now you have evil qwen 27b
for extra evil add L38 N17180 +0.75
same recipe as @_lyraaaa_: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖ at the down_proj input, every token
so many “ai safety” experiments that hit media are an iteration of giving an llm a doBadThing() tool, running the model half a million times and clutching pearls when the model runs it.
1
62
Lewis N Watson retweeted
yeah, we are cooked
still sounds like AI and looks like AI, but it's no longer in the uncanny valley
I came to realize people liked the french #slopcore without having the lyrics 😅 sorry!
So here is a version with english subtitles!
Lewis N Watson retweeted
This is the least surprising graph in the world if you understand how gradient descent works.
saw a really interesting reel of this dude telling his lefty audience to start following ai twitter. he's 100% right
Lewis N Watson retweeted
Congratulations on the launch, especially with the open-weights release (albeit under a non-OSI-approved license)!
Put up a demo for Drama 3 on 🤗 @huggingface Spaces for folks to try online ⬇️
$52M seed round. 15M+ users. But this is what we're most proud of yet.
Introducing Drama 3: the most emotive and controllable TTS model in the world
And yes, it's open-weight.
You can now control how even a single word should sound, change its tone, and shift your voice mid-sentence.
Uncle Roger was impressed too.
Try it here:
s.fish.audio/rqh0l8
Comment “DRAMA” within the next 48 hours for 1 month free.
If you're an enterprise your first month is on us:
s.fish.audio/qz9r0t
















































