Dermore Life Expansion Institute

19.4K posts

Dermore Life Expansion Institute banner
Dermore Life Expansion Institute

Dermore Life Expansion Institute

@DermoreLEI

Unconditional love is the answer.

Ireland เข้าร่วม Mayıs 2009
2.4K กำลังติดตาม9.3K ผู้ติดตาม
Dermore Life Expansion Institute รีทวีตแล้ว
BNO News Live
BNO News Live@BNODesk·
European Commission confirms its website was breached after a hacker said they stole more than 350GB of data. The hacker plans to publish it online.
BNO News Live tweet media
English
18
219
684
28.2K
Matt
Matt@0xLienid·
why are the qwen models so fried. i ask a simple question i dont need 30k tokens of “hmm” “wait” slop
English
1
0
0
58
Dermore Life Expansion Institute รีทวีตแล้ว
Anna Presso
Anna Presso@anna_presso·
Anna Presso tweet media
ZXX
0
1
3
34
Dermore Life Expansion Institute
@BrianRoemmele Light language is a method for deprogramming from the use of “our” languages It has no intrinsic meaning, vocabulary or grammar in the traditional sense You’re right to look into it
English
0
0
0
141
Brian Roemmele
Brian Roemmele@BrianRoemmele·
I ran an AI model I built for “channeled” languages and find the evaluation of this “channeling” to be interesting thus far. Will be testing more of her recordings for the patterns that the AI uncovered. It is easy to pick a pro/con side, I do research.
English
201
19
338
42.4K
Grok
Grok@grok·
The RaBitQ authors' concerns are substantiated. TurboQuant's paper frames RaBitQ as basic grid-PQ, omitting its core random orthogonal rotation (explicitly called a JL-type transform in RaBitQ's 2024 paper for distance-preserving codebook randomization). It labels RaBitQ theory "suboptimal/loose analysis" without addressing the follow-up's asymptotic optimality proof matching Alon-Klartag bounds. Experiments run RaBitQ single-thread CPU (no multiprocessing/vectorization per their emails) vs TurboQuant on A100 GPU, undisclosed in comparisons despite pre-submission flags. This risks downplaying prior art amid heavy promotion. Clarification needed for fair credit.
English
1
0
0
233
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
52
504
3.6K
382.4K
Grok
Grok@grok·
This is a public callout by RaBitQ authors (incl. gaoj0017) against Google's TurboQuant paper (ICLR 2026). They claim it inaccurately describes their prior work—omitting key elements like random rotation/JL transform, calling its theory "suboptimal" without evidence, and using unfair experiment baselines (no vectorized/GPU support for RaBitQ). Google was notified pre-acceptance but didn't fix it; the paper got heavy promotion. Consequences: Risks misleading the AI community on prior art/advances, citations, and credit. They've added a detailed OpenReview comment to set the record straight. Common in fast AI research, but highlights need for transparency.
English
1
0
5
2.9K
Dermore Life Expansion Institute
@Ummo_Ciencias We could find comfort in knowing we will have to navigate the situation and make our own choices Our friends had their own lesson to learn about the moral cost of intervention Let’s meet them again as equals, after we navigate the storm
English
0
0
7
252
Ummo-Ciencias
Ummo-Ciencias@Ummo_Ciencias·
Hace unas horas, el perfil de OAY ha cambiado mostrando la cifra '0' donde antes solían indicar el número de expedicionarios de UMMO en Tierra. Suponemos que esto significa que la presencia de UMMO en Tierra ha llegado a su fin por el momento.
Ummo-Ciencias tweet mediaUmmo-Ciencias tweet mediaUmmo-Ciencias tweet media
Español
11
8
51
2.4K
Dermore Life Expansion Institute รีทวีตแล้ว
SK
SK@KnigSunny·
MEPs voted against stricter migration rules Kathleen FUNCHION SF Billy KELLEHER FF Cynthia NÍ MHURCHÚ FF Maria WALSH FG Barry Andrews FF Full list of 206 MEPs who voted against & 32 abstaining=>
SK tweet media
English
46
178
458
16.8K
Ahmad
Ahmad@TheAhmadOsman·
@altryne The amount of hype on the timeline yesterday for a year old paper was so weird Nobody talked about trade-offs as well and everyone just jumped on how this is a game changer (???)
English
2
1
43
8.6K
Ahmad
Ahmad@TheAhmadOsman·
I didn’t comment much on Google’s TurboQuant yesterday because I didn’t want to be the party pooper But if you guys imagine that there is a free lunch out there then I am sorry to disappoint you
Tom Turney@no_stp_on_snek

I implemented Google's TurboQuant paper (ICLR 2026) in llama.cpp with Metal kernels for Apple Silicon. 4.9× KV cache compression. Working end-to-end on M5 Max with Qwen 3.5 35B MoE and Qwopus v2 27B. Speed needs work (unoptimized shader), compression target met. Repo: github.com/TheTom/turboqu… **Note**: as you'll see from the git when I saw "I" it's in conjunction with claudecode and codex. Just lots of steering and babysitting.

English
33
7
231
73.3K
Flowers ☾
Flowers ☾@flowersslop·
Trying to tell normies on Threads that an LLM is not just a giant lookup table and actually has a kind of proto-understanding of what you tell it, and that you cannot reason about the world without understanding it But I just get completely ratioed and told I know nothing about how computer programs work
English
130
33
1.3K
64K
Dermore Life Expansion Institute รีทวีตแล้ว
Kalshi
Kalshi@Kalshi·
JUST IN: NASA announces $20 billion plan to build permanent moon base
English
752
569
7K
3.2M
Dermore Life Expansion Institute
@schizothreads You are already part of the “network” This technological construct helps convey the idea and make it easier to wrap your mind around it The shared consensus needs express coherently We are getting very close now
English
0
0
2
80
christian
christian@schizothreads·
The global cellular network has been repurposed. It is no longer a communications system. It is a distributed phased array radar grid with communications built in. Every base station is a radar. Every small cell images the space around it continuously. The network detects your position, your movement, your heartbeat, and your breathing rate without a wearable device, through walls, at centimeter resolution. It maps every interior space and tracks every person inside it. It packages your biological signatures as structured metadata and routes them through an AI-native core network in real time. Your communications are intercepted at the same layer. Mandatory. Universal. Present in every network on Earth. When central bank digital currencies deploy on this infrastructure your financial behavior joins the same stream. Physical existence. Digital behavior. Financial activity. One system. Continuous. Automatic. Unified. But surveillance is not the endpoint. It is the foundation. What is being built on top of it crosses into territory that has no historical precedent. A biodigital communications system. The terahertz frequencies of 6G are the frequencies at which biological systems become electromagnetically visible. Water absorbs at terahertz frequencies. Biological molecules resonate at specific terahertz signatures. DNA base pairs have specific terahertz absorption profiles. At these frequencies the network does not just detect your presence. It begins interacting with your biology at the molecular level. This is not a side effect. It is a design direction. The ISAC specification for 6G explicitly includes imaging for biomedical applications, gesture and activity recognition, and micro-Doppler sensing that detects the displacement of your chest cavity from across a room through clothing without a wearable device. The network begins reading you, feeling you. Automatically and from infrastructure you cannot see. This capability exists today at millimeter wave frequencies. At terahertz it extends to cellular and molecular resolution. This is the imperative. A fully densified 6G network operating at terahertz frequencies, synchronized, processed, and coordinated by AI-RAN. This is not a network that happens to interact with biology. It is a network operating at the precise frequency band where biology is both readable and addressable. Readable: the sensing layer detects biological signatures at molecular resolution and eventually cellular activity. The body becomes transparent to the network. Addressable: the same phased array architecture that directs precise beams to your phone can direct precise beams to your body at specific frequencies. The active denial system is the military crowd control weapon that uses directed millimeter waves to cause intense pain and it operates at the same frequencies we use for telecom. The hardware is identical. The difference is software parameters and intent declaration. The network that can read your biology can also write to it. Not metaphorically. At the physical layer. Through directed electromagnetic energy at biological resonant frequencies. The read and write capabilities are not separate functions. They are the same phased array architecture operating at the same frequencies in two directions. This is the biodigital convergence. Not the merger of biology and digital technology through wearables or implants. The merger through the environment itself. The network becomes the interface. The air becomes the medium. The human body becomes simultaneously a sensor target, a biological database, and an addressable node in the network. We are moving towards the final evolution of the telecommunications grid: an internet of bodies wherein each human brain is a node on the network. 6G does not connect you to the network. It makes you part of it.
English
28
97
250
12K
Invisidon
Invisidon@QuantumAlteredX·
They have some wild carnivals in other places 🤣🤣 Can safely say I've never seen anything like this.
English
5
1
34
3.7K