Tomer Solberg

[Omer Shlomovits @OmerShlomovits]

Introducing HyperQuant: A new Lattice-based compression algorithm that significantly improves LLM key-value cache memory over TurboQuant and OCTOPUS, and with higher accuracy!
But it gets better - we apply the same compression technique to weights, outperforming Higgs!
Finally, (for the first time ever?), we show how to compress video DiT weights. demo on LTX-2.

Easy to integrate: we release a reference kernel code with the paper, under MIT license. This is a calibration-free, post-training quantization. Disclaimer: performance will get much better in next releases.

Links and key results in next tweet.

[Tomer Solberg @TomerSolberg]

@hikettei What about rescaling first and then applying the Hadamard rotations?

[Thomas Coratger @tcoratger]

CICLE V3.5: Sumcheck with Lambda Functions by @Ingo_zk

https://t.co/Fpe1UHG5K9

[Tomer Solberg @TomerSolberg]

@RenaudDUBOIS10 @OmerShlomovits The proof is only required if the ownership is disputed. Not every image needs a proof

[Tomer Solberg @TomerSolberg]

@sergey_nog @OmerShlomovits That's a good question. I think the number of samples you'll need increases exponentially with the bit length of the WM but It's definitely something to look into

[Tomer Solberg @TomerSolberg]

@RenaudDUBOIS10 @OmerShlomovits Meta's stable signature model discusses the robustness of the watermark. You'll have to distort the image by quite a lot to remove it as long as the decoder/extractor networks remain private

[Tomer Solberg @TomerSolberg]

This is T2 tower field
T2~ F2[x,y]/(x^2+x+1,y^2+xy+1)
Represent each number as a multilinear polynomial in x,y where the coefficients are the binary representation of the number.
So 1~1, 2~x, 4~y, 8~xy
This gives 6~x+y, and 2*6~x*(x+y)=x^2+xy=1+x+xy~1+2+8=11

[vitalik.eth @VitalikButerin]

2 + 2 = 0
2 * 6 = 11
3 / 5 = 9
3² + 8² = 11²
420 * 61779 = 1

It's the new math. If you use it, it becomes much easier to make proofs of things.

If you know, you know.

[Tomer Solberg @TomerSolberg]

@prasincs @kobigurk Actually no. Since this is a field it can't have 0 divisors so 2*6 is definitely not 0. Usually in GF(2^4) 2*6=12 (just look up galois field calculator), but I guess tower field representation is different than the standard one so it could be 11

[Ingonyama (Now MoonMath AI) @Ingo_zk]

We have successfully closed $21M in seed funding 🥳

The round was led by @IOSGVC, @walden_catalyst and @__geometry__

This is an important step on our mission to democratize #ZKPs by enhancing their speed, scalability, and cost-effectiveness, making them accessible to all

1/3

[Tomer Solberg @TomerSolberg]

@Antalpha_Labs Nice to hear that you found my write up useful! I considered adding Lasso, but I'm currently working on other projects and couldn't find the time

[Omer Shlomovits @OmerShlomovits]

more great educational content, by @TomerSolberg and @godstar28 👑
https://t.co/gpRlLq8kwx

[the Pit 🕳️ @jump_pit]

@Ingo_zk’s @TomerSolberg breaks down Fast Danksharding.

[Ingonyama (Now MoonMath AI) @Ingo_zk]

A Brief History of Lookup Arguments -
A new review paper!

1) uses of lookup protocols
2) plookup
3) cq, LIKE YOU HAVE NEVER SEEN BEFORE
4) evolution from plookup to cq

https://t.co/K4M6V0PQHn