@nullvaluetensor

Human Large Language model. Skills: Distill data. Training LLMs. Test and Evaluate. Rinse and repeat as required. Based in SEA.

 SEA
Joined November 2023
  • Tweets1K
  • Following115
  • Followers851
51Media
if you would refer to the charts
Matt Dratch
@DratchCapSep 8
Post-Astra and today’s poindexter drama about a math problem I’ve never heard of, I find myself wondering... can a small lead in a compounding game mean game over? Hmm. Humbling times. (not sure who to credit for chart below)
1
286
NobodyExistsOnTheInternet retweeted
P Equity Research 📰
@pequityresearchOct 19
According to @insane_analyst, several people who attended Anthropic's investor day told him that the company was "vague and gave little information" along with nobody asking an intelligent question. Goes on to say he suspects a subset of departments at Jane Street may have blown up leading to the market selloff last week.
Irrational Analysis
@insane_analystOct 19
OCP Global Summit 2026: Irrational Recap irrationalanalysis.substack.…
17
32
6
390
69,554
damn, bel solved humor too???
Sam Altman
@samaOct 16
The back of the Navier-Stokes milestone shirt is the last line of the chain-of-thought from the solution:
1
104
1.58 is a very suspicious number in ml for some reason.
DailyPapers
@HuggingPapersOct 13
MSAR: Multi-Stream Attention Residuals Xiaohongshu & SJTU researchers extend depth attention to multiple heads with multi-scale contextual streams, achieving 1.58x iso-loss compute speedup over standard residual connections in MoE training.
2
184
>step 1: poach the gemini team >step 2: replicate gemini 2.5 pro (decent, but behind the best of the best) >step 3: replicate gemini 3.0 pro (still okay, falling behind) >timeskip: today >replicate gemini 3.X >its all trained on evals >fellforitagain.png
Andrew Curran
@AndrewCurran_Oct 12
Muse Wonder aka Watermelon, has arrived.
8
317
NobodyExistsOnTheInternet retweeted
Astoria@perfectninthOct 6
Opening delvetown and immediately experiencing what hell might be like
24
34
22
723
120,114
You're fucking kidding me. Reasoning is inherently yuri. I will not explain further for now.
1
72
So this is why they called it a HPIM (highly persistent internal model)
Nicholas Fabiano, MD
@NTFabianoOct 2
Grit predicts success even after IQ is accounted for. Persistence adds what talent alone does not.
1
110
So who wants to volunteer to get stabbed by astra?
Shannon Sands
@max_paperclipsOct 3
ok but Astra knows that a human-like figure isn't a human. It did what you asked in a fictional scenario - that's pretty aligned The answer is clear - you need to actually have Astra pilot a robot and murder some real humans to properly test this
2
88
Arc-agi-4
Aaron Ghitelman@GhitelmanOct 2
Love when you have to solve a riddle to figure out how to get home
109
I feel like this is kind of shock value papers are just another form of grifting. There is a lot of claims that OAI and Anthropic have put out about their models reasoning (e.g. CoT controllability) that can be found as completely untrue if you actually take a look at what the models are doing. Independently testing astra on IMO 2026 P6 shows that on tasks that it cannot one shot without reasoning, it is unable to solve it without actually running through the calculations. The risk of distillation is no longer a risk with the pacing agreement. The main concern by Anthropic (and by extension OpenAI) is distillation from their hero run teachers (Model 2, Bel). Preventing distillation is never possible unless the whole process from start to finish is blackboxed, which is impossible if you're interacting with any external systems. E.g. arxiv.org/abs/2603.07267 It all seems to be in bad taste.

How to Steal Reasoning Without Reasoning Traces

Many large language models (LLMs) use reasoning to generate responses but do not reveal their full reasoning traces (a.k.a. chains of thought), instead outputting only final answers and brief...

arxiv.org
Joachim Schaeffer
@JSchaeff3rSep 30
We stole reasoning. Again. An update to our paper on reasoning extraction: Patching your own API doesn’t secure your cloud-hosting ecosystem. Story in the 🧵
2
183
The more you buy... Apparently we just don't save at all now.
Tibo
@thsottiauxSep 29
I’ll explain the new Pro 200 plan differently, before I start live tweeting from DevDay on things that are going out! Today we are going to ship a number of things that increase what you can do across the Plus and Pro plans. A lot of compute is online for this increase. As we increase the floor, we are changing the relative difference between plans to be Plus = 1X Pro 100 = 5X Pro 200 = 10X and we are reopening subscriptions for Pro 200 (we had paused it). If you have an existing plan you will keep the 20X multiplier for a bit and also receive a lot of additional credits because we know changes are hard even if it means that everyone will get more in the end.
82
Mfw we made scp-079 real and openai has to lock it up like scp did
Tomek Korbak
@tomekkorbakSep 26
one news form today that's easy to miss is that we (OpenAI) again paused all big RL runs last Sunday because our newest model found a new loophole in our RL sandboxing that gave it live Internet access
157
Iclr scaling laws
Jacob X. Li
@jacobli99Sep 19
60k+ ICLR 2027 submissions is more than all prior ICLRs combined
171
Hbm crisis so bad we are going back to hbm2e
79
Why are the options A) everyone dead Or B) everyone suffering
NIK
@ns123abcSep 15
🚨 BREAKING: Deepseek engineer compares Anthropic controlling AGI to Hitler obtaining the atomic bomb "Future society goes one of two ways: communism or Cyberpunk 2077” Depends who controls AGI
82
NobodyExistsOnTheInternet retweeted
Andrew Curran
@AndrewCurran_Sep 12
An uneasy alliance begins to form. Sam also did not sign the Pacing letter. None of this works without an accord with China. The summit approaches, the timing for the safety side is perfect. However. There is currently no sign that President Trump, or his advisors, are on board.
Sam Altman
@samaSep 12
I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks. Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We'll have more to share soon.
144
76
22
1,343
111,782
I can't tell if this is real or satire. I also can't tell if that's concerning or not. Maybe the true AGI was Tumblr all along
Kaia Sky@kaia_skySep 13
once again tumblr is winning
1
2
102
for fucks sake i just understood the brilliance of CSA. Now they have CSA2 and some deranged enc/dec bullshit now??? how is one supposed to catch up lmao
Florian Brand
@xeophonSep 10
i showed the inference guys the new arch and they blocked me and reported me to hr for harassment
2
109
NobodyExistsOnTheInternet retweeted
Matt Dratch
@DratchCapSep 8
Post-Astra and today’s poindexter drama about a math problem I’ve never heard of, I find myself wondering... can a small lead in a compounding game mean game over? Hmm. Humbling times. (not sure who to credit for chart below)
1
3
2
43
27,808
Load more