Vraj Shah

Staff Research Scientist @IBMResearch. PhD @ucsd_cse. BTech @IITIOfficial.

[Sara Rosenthal @seirasto]

Introducing MTRAG-UN — a new benchmark for the UN parts of multi-turn RAG!

666 tasks, 2,800+ conversation turns, 6 domains — including two new enterprise corpora

📄 Paper: https://t.co/wXgzyEsdjv
💻 Benchmark: https://t.co/qLZvVbQbhV

collab: @vpshah95 @lucian_popa_us

[Sara Rosenthal @seirasto]

🌟New Benchmark! 🌟

Do you work on RAG? Are you interested in Multi-Turn conversations? Very excited to share the new MTRAG benchmark we have released!

Data: https://t.co/KtJQgtB5Uj
Paper: https://t.co/QNRccmrEV5

@danish_c @kpfadnis @chulaka_g @vpshah95 @lucian_popa_us

[Vraj Shah @vpshah95]

@TweetAtAKK Thank you so much @TweetAtAKK. This would not have been possible without your support.

[Vraj Shah @vpshah95]

@shashank_r12 @DimitrisPapail Congratulations Shashank!

[Vraj Shah @vpshah95]

@TweetAtAKK @UCSanDiego @scnakandala @ucsd_cse Thank you @TweetAtAKK

[Vraj Shah @vpshah95]

@MadelonHulsebos @TweetAtAKK @ucsd_cse @UCSDJacobs @UCSanDiego Hi @MadelonHulsebos thank you so much for the wishes! Hope that you are doing well.

[Vraj Shah @vpshah95]

@GarrisonHess @TweetAtAKK @ucsd_cse @UCSDJacobs @UCSanDiego Thank you so much, Garrison!

[Vraj Shah @vpshah95]

@niviksha @TweetAtAKK @ucsd_cse @UCSDJacobs @UCSanDiego Thank you so much for the wishes, Venkat!

[Vraj Shah @vpshah95]

@TweetAtAKK @IBMResearch Thank you so much Arun for everything! This thesis would not have been possible without your support. Very fortunate to have you as my advisor and I can't thank you enough.

[Ryan Marcus @RyanMarcus]

As I finished up my academic interviews, I had some time to reflect on the most useful advice that I got or wish I had gotten for the CS academic job market:

1) You need an excellent 45m job talk. Don't believe anyone who says you have any longer than that.

[Vraj Shah @vpshah95]

@MadelonHulsebos @TweetAtAKK @adityagp @arnabdotorg Very nice and interesting work! @MadelonHulsebos Happy to discuss this further and take it up offline.

[Matthias Boehm @matthiasboehm7]

The DEEM workshop (DEEM: Data Management for End-to-End Machine Learning) @2022Sigmod goes into another round. So don't miss out on a hopefully hybrid workshop with many unexpected discussions, and submit your papers by Mar 01:
https://t.co/2xlcPdYsSW

[Arun Kumar @TweetAtAKK]

1/7: 🚨ATTENTION!🚨

A panel debate you DO NOT WANT TO MISS is coming soon to @SIGMOD_2021!

Thank you @joe_hellerstein, @ihabilyas, @sarahcat21, Luna, Felix, & Dr. A. Diaboli for joining!

ML/AI hell hath no FURY like DB folks scorned! This is WAR! 🎆

Spread the word. A 🧵👇

[Anmol @anmolarora6699]

We've launched a new forum for people to find resources related to Covid-19. Team & Volunteers at @DocVitaOfficial check for information and bring it all together in one place for you to access -

https://t.co/t8YmCPV0mv

[Sibin @sibinmohan]

I’m seeing a massive disconnect (more so than usual) between the two twitter worlds I usually inhabit — the US and India. It is important to bridge the two now for a very important reason: #CovidIndia

Please read the whole thread and amplify.

THREAD

1/17

[Arun Kumar @TweetAtAKK]

Delighted to share this overview of our strong & growing Database Lab! 🙌 @ucsd_cse @HDSIUCSD

Video: https://t.co/FDkawmSNpd
PDF: https://t.co/NnS1qPXqAX

Learn about the group's faculty, history, current research projects & students, & the eclectic data tech ecosystem in town.

[Vraj Shah @vpshah95]

@TweetAtAKK @ucsd_cse @UCSDJacobs Thank you so much, Arun. I am very grateful, would not have been possible without your support.

[Vraj Shah @vpshah95]

@TweetAtAKK Thank you, Arun for all your support.

[Andreas Mueller @amuellerml]

Just saw an amazing talk by @TweetAtAKK on his work on detecting feature types for ML with the ML data prep zoo: https://t.co/vOjeALQwAb
Excited to use it to benchmark how dabl is doing and improve its type detection! Talk will be available here soon: https://t.co/0uYGTy2QMK

[Juan Sequeda @juansequeda]

My #sigmod2020 Trip Report.
Takeaways:
1) Gap between real world and academia on data integration
2) Academia is starting to study data catalogs
3) More interest in non-SQL interfaces to access data
4) Graphs continue to gain more industry traction

https://t.co/3fukdqt31N