Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise · HackerPapers discussions

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise — mentioned by GaggiX

Tue, 14 Mar 2023 12:25:54 GMT

I wrote a response yesterday but did not post it or send it, ops. I still don't understand the problem, if you ask model trained on a noise pattern "trees" for a forest it will still give you a random forest, that's what it was trained on, also: https://arxiv.org/abs/2208.09392, to see the diffusion process applied to processes other than Gaussian noise.

Paper and discussion history. A mention is not an endorsement.

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise — mentioned by astrange

Sat, 07 Jan 2023 07:26:38 GMT

This claims to explain diffusion models from first principles, but the issue with explaining how they work is we don't know how they work. The explanation in the original paper turns out not to be true; you can get rid of most of their assumptions and it still works: https://arxiv.org/abs/2208.09392

Paper and discussion history. A mention is not an endorsement.

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise — mentioned by krackers

Tue, 04 Oct 2022 21:05:17 GMT

I think it might be this paper [1] succintly described by the author in this twitter thread [2] [1] https://arxiv.org/abs/2208.09392 [2] https://twitter.com/tomgoldsteincs/status/156250381442263040...

Paper and discussion history. A mention is not an endorsement.

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise — HN submission

Mon, 29 Aug 2022 17:38:58 GMT

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise

3 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise — HN submission

Mon, 22 Aug 2022 03:42:45 GMT

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise

9 points; 2 discussion replies on HN. Paper and source history. A submission is not an endorsement.