No I first was and then I wasn’t when I saw that paper and all the congrats in the replies . (How do we expect LLMs to ever solve text comprehension)
@rasbt
-
Congratulations on Real XAI Research Work
By
–
Haha I am just realizing that you are working on real XAI, not xAI)! In that case, real non-sarcastic congrats!! We need more of this!
-
KNN Classification Works Through Majority Vote Among Neighbors
By
–
Yes, I assume this should work because the compute the label based on majority vote among the nearest neighbors.
-
Nearest-Neighbor Approach on Normalized Bag-of-Words Vectors
By
–
Thinking about this a bit more, I think nearest-neighbor on normalized bag-of-words vectors would probably also perform well here. Because the count vectors of two similar documents are similar just like the compressed + concatenated compressed docs are similar.
-
Compression of Concatenated Identical Texts Size Analysis
By
–
Yeah, I think if two identical texts are concatenated, the size of the compressed concatenated file should match the size of the unconcatenated files right?
-
Gzip and kNN Outperform Transformers for Text Classification
By
–
Gzip + kNN beats transformers on text classification. (Gzip as in good old zip file compression)
-
Sarcastic critique of AI alignment solution timeline claims
By
–
Congrats! You should join forces with the other company who just announced that they will solve AI alignment in <4 years. So you can both work on “solving the universe” once your are done with these more near-term projects.
-
Self-Supervised Learning for LLM Pretraining and Evaluation Methods
By
–
We use self-supervised learning to pretrain LLMs (e.g., next-word prediction). Here's an interesting take using self-supervised learning for evaluating LLMs: https://
arxiv.org/abs//2306.13651
Turns out, there's correlation between self-supervised evaluations & human evaluations. -
Miniforge: The Go-To Package Manager for M1 Mac Python
By
–
Yes, same! When I upgraded to an M1 mac a few years back, miniforge was the only one supporting Python for Apple Silicon , and I've been using it ever since!
-
LLMs as Coding Companions: Automating Low-Level Tasks with Human Guidance
By
–
Yeah, I think LLMs will automate most of the low-level programming tasks. They will get you +90% of the way there. But I still see it more as a coding companion that needs human interaction and guidance.