Deep Learning @DeepLearning18

Deep Learning - Life School Joined June 2019

Tweets

10
Followers

136
Following

13
Likes

4

Andrew Ng @AndrewYNg

3 years ago

New short course on sophisticated RAG (Retrieval Augmented Generation) techniques is out! Taught by @jerryjliu0 and @datta_cs of @llama_index and @truera_ai , this teaches advanced techniques that help your LLM generate good answers. Topics include: - Sentence-window retrieval, which retrieves not just the most relevant sentence, but a window of sentences around it for higher quality context. - Auto-merging retrieval, which organizes your document into a hierarchical tree structure, where each parent node's text is split among its child nodes. Based on the relevance of the child nodes to a user query, this lets you better decide whether the entire parent node should be provided as context to the LLM. - Evaluation methodology for separately evaluating the quality of the key steps of RAG (context relevance, answer relevance, groundedness) so that you can perform error analysis, identify which part of your pipeline needs work, and tune components systematically. Please check out the course! deeplearning.ai/short-courses/…

57 442 3K 655K 2K

View Details

Deep Learning @DeepLearning18

3 years ago

@arai_ksaa @KSGAFAL من حق المشاركين عليكم تبليغهم بتاريخ للإعلان عن الفائزين! قد تبدو تشويق وإثارة في أعين المنظمين ولكن المشاركين يترقبون إعلانكم مع كل يوم جديد! ما المانع من ذكر أسماء الفرق الفائزة وارجاء ترتيبها ليوم محدد ومعلن!

1 0 1 139 0

View Details

David Mimno @dmimno

4 years ago

The Text as Data Conference (TADA) 2022 will be happening at Cornell Tech on beautiful Roosevelt Island in NYC Oct 6-7. Abstract submissions are due July 18. Please RT!

3 109 244 0 38

View Details

Dan Jurafsky @jurafsky

5 years ago

Here's the December 2021 release of draft chapters for Speech and Language Processing, just in time for Jim and I to wish you all a Happy New Year! Enjoy! web.stanford.edu/~jurafsky/slp3/

7 149 616 0 98

View Details

Deep Learning @DeepLearning18

4 years ago

@irinablok @_arohan_ 🗣🤖

0 0 0 0 0

View Details

Basil Mustafa @_basilM

4 years ago

Beep beep! Introducing LIMoE, the Language Image Mixture of Experts: a single model, processing both modalities for contrastive image-text modelling. Cruises straight to 84.1% 0shot ImageNet accuracy without any modality-specific architectures or pre-training. (1/10)

3 38 193 0 53

View Details

rohan anil @_arohan_

4 years ago

Just remembered again how true this with learning the LR literature, there is so much compute spend in finding meta-hyper-parameters (that’s hidden away) that results look great in paper but fails on new problem.

Ben Poole @poolio

4 years ago

@SussilloDavid "We present a hyperparameter-free method..." def beautiful_method(): return method(lr=3e-4, alpha=0.13, gamma=pi/9.1)

1 0 29 0 1

1 1 20 0 0

View Details

Darek Kłeczek @dk21

4 years ago

How train the best performing NLP model? 3 techniques to experiment with. 🧵

2 17 90 0 48

View Details

Stanford NLP Group @stanfordnlp

4 years ago

The SERAC model for small edits to large neural networks by @_eric_mitchell_ Charles Lin @ABosselut @chrmanning @chelseabfinn is the bottom item in @DeepLearningAI_’s The Batch this week. We’re not sure why this week—it must be a slow AI news week. 😭 deeplearning.ai/the-batch/issu…