Publications, books, and web pages I've been reading or am planning on reading.
I've been trying to level up recently on ML, LLMs, NLU, etc. and whenever I read a paper, I feel there are ten others I should read as well :) . This repo is to better track what I've read and what I want to read and jot some learnings along the way.
I also want to give this Learning in Public thing a shot. Let's see how it goes!
| Paper | Read Date | Last Revise Date | Notes |
|---|---|---|---|
| Evaluating Large Language Models Trained on Code | 2023-03-12 | ||
| Understanding HTML with Large Language Models | 2023-03-12 | 2022-10-08 | Notes |
| Multi-Task Sequence to Sequence Learning | 2016-03-01 | ||
| Emergent Abilities of Large Language Models | 2022-10-06 | ||
| BLOOM: A 176B-Parameter Open-Access Multilingual Language Model | 2022-12-11 | ||
| Finetuned Language Models Are Zero-Shot Learners | 2022-02-08 | ||
| LLaMA: Open and Efficient Foundation Language Models | 2023-03-27 | ||
| Training language models to follow instructions with human feedback | 2022-03-04 | ||
| HTLM: Hyper-Text Pre-Training and Prompting of Language Models | 2021-07-14 | ||
| Environment Generation for Zero-Shot Compositional Reinforcement Learning | 2022-01-21 | ||
| BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding | |||
| LaMDA: Language Models for Dialog Applications |
| Read Date | Resource | Notes |
|---|---|---|
| 2024-01-30 | Bloom Filters by ByteByteGo | Gives a decent intuition |
| 2024-01-30 | What are Bloom Filters? | Not the best example. prev vid was better |
| 2024-01-30 | Advancing Spark - Bloom Filter Indexes in Databricks Delta | Interesting, but more about delta than spark, as the title implies |
| The Case for Learned Index Structures | ||
| Optimizing Learned Bloom Filters by Sandwiching |
| Read Date | Resource | Notes |
|---|---|---|
| Permute, Quantize, and Fine-tune: Efficient Compression of Neural Networks | ||
| 2024-01-30 | How We Scaled Bert To Serve 1+ Billion Daily Requests on CPUs (Roblox) | Interesting. Always nice to read actual case studies. I'd like to see how ONNX compares to their benchmarks. |
| Read Date | Post | Notes |
|---|---|---|
| 2024-01-30 | How we reduced our text similarity runtime by 99.96% (Microsoft) | I skimmed through it. Seems interesting and worth a reread |
| 2024-01-30 | How Roblox Reduces Spark Join Query Costs With Machine Learning Optimized Bloom Filters | I wonder if this can be applied to other use cases too and not just fact/dim tables. Interesting read. |
| Post | Notes |
|---|---|
| Using machine learning to index text from billions of images (Dropbox) | Curious abouth the OCR/PDF text extraction part here. Need some caffiene in me to read this. |