Md Zia Ullah

Lecturer/Asst. Professor at Edinburgh Napier University

[Md Zia Ullah @zuacsea]

Call for paper: CIRCLE 2022 | 2nd edition of the Joint Conference of the Information Retrieval Communities in Europe, July 4-7, 2022, Samatan, Gers, France
https://t.co/nTXERUtaI2

[LlamaIndex 🦙 @llama_index]

This weekend, we’re providing a definitive set of tutorials on how to build GraphRAG, step-by-step.

First, check out this video by @fahdmirza on implementing the core components of GraphRAG using an in-memory implementation:
1. Extract entities and relationships using LLMs
2. Partition graph into communities and generate community summaries
3. Query all communities and synthesize into a final answer.

Then, extend this initial in-memory implementation by storing your property graph in a @neo4j graph database. Big shoutout to @ravithejads for the notebook

Video: https://t.co/unDAvDrRCv
V1 Notebook: https://t.co/eJRvyJSe6i
V2 notebook (with @neo4j): https://t.co/B91rH34PIU

Source GraphRAG paper by Edge et al. (+ image credits): https://t.co/bfhihoDYEP

[Nando de Freitas @NandoDF]

The Llama 3 paper is a must-read for anyone in AI and CS. It’s an absolutely accurate and authoritative take on what it takes to build a leading LLM, the tech behind ChatGPT, Gemini, Copilot, and others.

The AI part might seem small in comparison to the gargantuan work on *data* and *scale engineering*.

I hope professors in distributed systems, high performance computing, algorithms, databases, HCI, etc use it as an example of bleeding edge CS in their classes. So many exciting open problems!

@UBC_CS @CompSciOxford @berkeley_ai @Cambridge_Eng @WitsUniversity @NSERC_CRSNG @NSF @ERC_Research @UKRI_News

[Negar Arabzadeh @NegarEmpr]

Not knowing how to say “NO” caused me being swamped with 21 papers to review for #CIKM conference (different tracks) 😵‍💫🤯🥴
Even dedicating 2 hours to each paper, is going to consume a regular workweek worth of hours (not the never-ending PhD student week though) 😂
#phdlife

[Md Zia Ullah @zuacsea]

Welcome to CIRCLE2022 project session https://t.co/reMYRsW9UF

[Addy Osmani @addyosmani]

The "Software Engineering at Google" book is now free to read online: https://t.co/cqX79uZNFK. Great insights on improving code quality & scaling engineering teams.

[MIT CSAIL @MIT_CSAIL]

800 free computer science classes you can take online right now: https://t.co/RURjy6O01h

[Md Zia Ullah @zuacsea]

@maliannejadi @UvA_Amsterdam @irlab_amsterdam Congratulations! Super very happy for you

[Jessica Rodrigues @rodriguesjm6]

Want to improve your academic writing?

Here are 5 tweets full of excellent writing tips.

Because it’s about time I start sharing all the great tweets I bookmark. @AcademicChatter #AcademicTwitter

[Hideo Joho @hideo_joho]

Special Issue on Fair and Explainable Information Access Systems for Social Good - Call for papers - Information Processing & Management - Journal - Elsevier, Due: Oct 15, 2021 https://t.co/W1HNEVJIeL

[Fermat's Library @fermatslibrary]

Here's a visual proof that
sin2x = 2sinxcosx
cos2x = cos²x−sin²x

[Magdalena Kaiser @mag_kaiser]

Code, slides and video for our SIGIR long paper "Reinforcement Learning from Reformulations in Conversational QA over KGs" are now available:
Code: https://t.co/9AHVA1N4HO
Slides: https://t.co/jtfiH37mkG
Video: https://t.co/1tChufszMI
@RishirajSahaRoy #sigir21 #questionanswering

[Olivier Bachem @OlivierBachem]

Struggling to make your #NeurIPS2021 submission look neat and fit on 9 pages? Try putting \looseness=-1 at the beginning of paragraphs that you want Latex to format one line shorter. It's like \vspace{-xmm} but better :D

[Thomas Dehaene @TDehaene]

Me starting a new #NLP side project and listing up all the tools I wanna try out!

@huggingface @deepset_ai @explosion_ai #AI

[Jure Leskovec @jure]

Stanford CS224W Machine Learning with Graphs. Lectures 2 and 3 posted! Feature-based learning (Motifs, Graphlets, Graph Kernels) and Node Embeddings (DeepWalk, Node2Vec, Anonymous Walks).
Videos: https://t.co/zwFPXUSxvY
Syllabus: https://t.co/gkD71I61xW

[Santiago @svpino]

If you are looking to get a background in math before starting with machine learning, here is all the material you need covering the following topics:

• Probabilities & Statistics
• Linear Algebra
• Multivariate Calculus

More than enough to get started.

↓ 1/7

[Fernando Diaz @841io]

I suspect a lot of students are taking their first information retrieval course this semester, which, at this point, includes PageRank as a core concept. Let's discuss. 🧵 1/n

[Craig Macdonald @craig_macdonald]

Parts 1 & 2 of the #BoW2B tutiorial @2021Ecir covered PyTerrier basics and learning-to-rank. Missed it? Catch up w/ slides and notebooks from https://t.co/jwfBYK1oeS.
In 10 mins, @macavaney will start Part 3 discussing neural re-rankers.

[Md Zia Ullah @zuacsea]

@arjenpdevries @lintool Finetuning vs. training? To my understanding, finetuning means you keep the pre-trained weights for the first few layers (as generic) and update the weights of latter layers (as downstream). Training means you update the weights of all layers starting from random or pre-trained.

[elvis @omarsar0]

It still amazes me how this formula, implemented in a few lines of Python code, is powering some of the latest breakthroughs in deep learning.