Sanket Daru

[Sanket Daru @sanketdaru]

@srdomcogbbcr @drmmumbaicr @RailwaySeva What is being done to not let the schedule of the train slip further? Every subsequent station is now showing greater delays than that at Lonavala.

[Sanket Daru @sanketdaru]

@WesternRly @Central_Railway train number 09420 (origin 30th Aug) is running late and is getting further delayed..! There are senior citizens travelling and these delays are causing severe inconvenience. Can you please assist with this matter?

[Sanket Daru @sanketdaru]

@RailwaySeva @drmmumbaicr Shared details via DM

[Sanket Daru @sanketdaru]

@drmadiwr @RailwaySeva Just now someone came over to resolve the situation. Thank you.

[Sanket Daru @sanketdaru]

@RailwaySeva train 22920, coach B10, seat 71. The toilets are overflowing, filthy and smelling very badly. Kindly arrange a cleaning immediately.

[Sanket Daru @sanketdaru]

@drmadiwr @RailwaySeva Thank you, however no official is attending the complaint although the train is stopped at Surat.

[LangChain @LangChain]

⭐️ Multi-Vector Retriever for RAG on tables, text, and images ⭐️

Seamless question-answering across diverse data types (images, text, tables) is one of the holy grails of RAG.

We’re releasing three new cookbooks that showcase the multi-vector retriever to tackle this challenge.

We released the multi-vector retriever back in August w/ a simple idea:

1/ embed a doc reference (e.g., summary) that is optimized for search

2/ but retrieve the raw doc (table, text, image) to give complete context for LLM answer synthesis

Using @UnstructuredIO to parse images, text, and tables (e.g., from pdfs) ...

.. the multi-vector retriever enables RAG on semi-structured data w/ table summaries -
https://t.co/Vmxcifa81A

.. we extend the idea to images, using LLaVA-7b (c/o @imhaotian) to produce image summaries -
https://t.co/PufDyCGM9J

... and this full RAG pipeline can be run laptop w/ llama.cpp c/o @ggerganov, @ollama_ai, @nomic_ai embeddings, and @trychroma:
https://t.co/yXxlxGkjbq

Blog:
https://t.co/U5tymTXdYY

[Akari Asai @AkariAsai]

Introducing Self-RAG, a new easy-to-train, customizable, and powerful framework for making an LM learn to retrieve, generate, and critique its own outputs and retrieved passages, by using model-predicted reflection tokens.
📜: https://t.co/CbIdZeAmcR
🌐: https://t.co/j1kRrrS3Go

[Sanket Daru @sanketdaru]

@tata_neu had a harrowing experience with your customer “non” service. Two gentlemen were playing ping-pong sending me from one person to another claiming that the other didn’t do their job right! Also sent an email to follow up which has not been responded to in 3 days!Pathetic.

[Shreya @shreyamatsharma]

A slightly different quirk about #ICCWorldCup2023 today for you:

Do you know what these symbols mean during the visuals we see during world cup games?

There are 9 in total and you would have seen them use everywhere from scorecard visuals to logos to branding.

A thread 🧵

[Sanket Daru @sanketdaru]

@RailMinIndia yesterday night nobody opened doors B7 of 22498 PNR 4151395840 SLI 0240 HRS.Single lady travelling, no help.She had to run many compartments ahead to board from any open door. TC, staff, everyone were fast asleep least bothered to perform their duties! Appalling!

[Avi Sil @aviaviavi__]

#acl2023 findings paper for folks working on retrieval leaderboards- Read on:
✅ We show multi-dimensional tradeoffs e.g. quality , latency & cost (instead of just F1)
✅ Metrics that include concrete efforts e.g. DynaScore.

-- Code in PrimeQA: https://t.co/R6GbTTkY90

[Abid Hassan 🇮🇳 @abidsensibull]

It has been a month since I have been on Social Media. No Twitter, no Instagram, no FB (is there anyone there anyway?) etc. Also did not watch any OTT. And the effect has been quite profound for me. Here are my learnings:

[Sanket Daru @sanketdaru]

@VikramAinchwar @jagograhakjago @zomato Ice cream ko kaun NAA karta hai dost? Waise ads naa dekhne ke liye premium plan liya tha aur jale pe namak, distasteful ads woh bhi bas achanak se shuru ho jaati hai. Koi 5vi paas developers ko bitha ke kaam karvaaya ho waisa feel aata hai. @SonyLIVHelps #DoesNotHelp

[Sanket Daru @sanketdaru]

@SonyLIVHelps @SonyLIV why are ads shown on recorded #SharkTankIndia episodes? I am a premium “paying” customer! This falls under “unfair trade practices” If you want to run ads, refund my subscription amount! #unfair #cheating

[Sanket Daru @sanketdaru]

At least have the courtesy to respond to your paying customers! Is that too much to ask for?Or by remaining silent you are indicating that “pay us and make a fool out of yourself” #cheating #looting in the name of #SharkTankIndia #consumerprotection needed @jagograhakjago

[Sanket Daru @sanketdaru]

In your official apps FAQs, there is no mention of ads being served on reality show replays! Clearly a malicious intent on your part to earn more by saying something and showing something else! A case of #cheating and #BreachOfTrust CC: @jagograhakjago

[Sanket Daru @sanketdaru]

@HDFCBank_Cares You can request patience for a week or two. My original complaint was raised on 7th January and since then all I have got is insensitive and illogical comments from @HDFC_Bank. I was once a vocal supporter of yours, definitely not anymore. The quality of support now is abysmal.

[Sanket Daru @sanketdaru]

@HDFCBank_Cares on 7/01 raised 45836057 unresolved till 11/01.Escalation 47873687 raised with SLA of 18/01.Agent asked for 1 more day, yet no resolution!Unbecoming behaviour of repeatedly referring me by someone else's name is height of unprofessionalism.Need further escalation!!

[Sanket Daru @sanketdaru]

@HDFCBank_Cares Varun, do you read before you tweet? I already informed you that I had responded to your email in my last tweet! This is how you care for your customers? @HDFC_Bank

[Sanket Daru @sanketdaru]

@HDFCBank_Cares At least cross check your facts before writing on public forum Varun. That email you’re talking about, responded to it at 1843 HRS IST. Waiting for your team’s response on the same. I hope you will be thorough in solving the problem.