Diffuser: Efficient Transformers with Multi-hop Attention Diffusion for Long Sequences
— AK (@_akhaliq) October 24, 2022
abs: https://t.co/e0ZSrPRoeH pic.twitter.com/V9STZ7nmQU
Diffuser: Efficient Transformers with Multi-hop Attention Diffusion for Long Sequences
— AK (@_akhaliq) October 24, 2022
abs: https://t.co/e0ZSrPRoeH pic.twitter.com/V9STZ7nmQU
Large Language Models Can Self-Improve
— AK (@_akhaliq) October 24, 2022
abs: https://t.co/Hjp5AfvFD3
approach improves the general reasoning ability of a 540B-parameter LLM (74.4%→82.1% on GSM8K, 78.2%→83.0% on DROP, 90.0%→94.4% on OpenBookQA, and 63.4%→67.9% on ANLI-A3) pic.twitter.com/yYW7e2WsPB
It's always exciting to add a new method to the deep tabular list: https://t.co/pWOAwqiTS0.
— Sebastian Raschka 📚 (@rasbt) October 23, 2022
Just read through the paper. It's an intriguing fresh take on deep learning for tabular data, combining approximate Bayesian inference and transformer tokenization. [1/6] https://t.co/iJeYGFC9dG
Accurate. pic.twitter.com/7AXTPP7Li7
— Bojan Tunguz (@tunguz) October 22, 2022
Wow, below is the 2nd paper shared today on deep tabular methods!
— Sebastian Raschka 📚 (@rasbt) October 21, 2022
The deep learning from tabular data research field is on fire today 🔥!
(PS: And don't forget about diffusion models for tabular data: https://t.co/CaEWyMyqnK 😁)
(PPS: My reviews will follow soon 😊) https://t.co/xupH4Znh3e
A meta-learned transformer for tabular data: https://t.co/3rlNtJa2hc I was waiting for this to happen, and I'm pretty convinced that's where ML will go. From Frank Hutter's (imho legendary) lab!
— Andreas Mueller (@amuellerml) October 21, 2022
🤔 You probably never heard of MMRotate
— @farid@sigmoid.social (Mastodon) (@ai_fast_track) October 21, 2022
Here are some reasons why you should be familiar with:
• MMRotate is an open-source toolbox for rotated object detection based on PyTorch
• Like the awesome MMDetection library, It is part of the @OpenMMLab project (18 libraries 🤯) pic.twitter.com/yd4VhmVkc7
MovieCLIP: Visual Scene Recognition in Movies
— AK (@_akhaliq) October 21, 2022
abs: https://t.co/WaF11rNcX1
project page: https://t.co/ZTK9PYWmcW pic.twitter.com/KyPsp3DkNt
DiffEdit: Diffusion-based semantic image editing with mask guidance
— AK (@_akhaliq) October 21, 2022
abs: https://t.co/9pzsPwiU1K pic.twitter.com/elAYmnMh2w
😨 Training an Object Detection Model is a very challenging task and involves tweaking so many knobs
— @farid@sigmoid.social (Mastodon) (@ai_fast_track) October 20, 2022
Here is an exhaustive 🎁 tips & tricks list 🎁 that you could use to boost your model performance
🧵 pic.twitter.com/sOvEUhCCwg
RankT5: Fine-Tuning T5 for Text Ranking with Ranking Losses
— AK (@_akhaliq) October 20, 2022
abs: https://t.co/j8i1CgQrYb pic.twitter.com/qG1VqFWwEb
A Unified View of Masked Image Modeling
— AK (@_akhaliq) October 20, 2022
abs: https://t.co/Kw4zgbpgXY pic.twitter.com/WS9RgFU4NJ