-
As detailed in the paper "Pseudo Constant Time Implementations of TLS
Are Only Pseudo Secure"
https://eprint.iacr.org/2018/747
-
Hi there!
A colleague of mine (h/t @maxammann) shared your S&P paper with me. I think path validation benefits hugely from formal validation, and I was very excited to see your work in this space!
…
-
[P3136R0](https://wg21.link/p3136r0) Retiring niebloids (Tim Song)
-
Tune hyperparameters / match implementation details / fix bugs until we replicate the performance of reference implementations of algorithms. I'm not concerned about an exact match -- if we do about a…
-
[P2767R0](https://wg21.link/p2767r0) flat_map/flat_set omnibus (Arthur O'Dwyer)
-
1. Remove hard-coded trivial actions, validate that network learns them.
2. Experiment with co-learned H. If you see problematic circling again, do:
3. Separate H-training
4. Implement CFR (I sent …
-
Hi, after reading your paper and studying the code, I don't understand why VisionTransformerForMaskedImageModeling have two implementations of the encoder (respectively encoder and teacher model). Why…
-
I noticed that in your implementation you've clamped the values of the `weight_norm` to min of `0` and max of `10`. I have seen this `10` before in other implementations and noticed that this comes f…
-
hey, was just reading this paper http://www-users.cs.umn.edu/~lv/FBC.pdf and looked for implementations but didn't find any. it proposes some pretty interesting space saving capabilities that go beyon…
-
Hi,
I am using the word embeddings that Lample has used in his paper. However, I got lower/better results than Lample when I use your implementation to train and test the BiLSTM-CRF model. Do you kno…