Dilip Thomas Ittyera
@diliptiMaking GenAI Reliable #AugmentedIntelligence
Similar User
@avinashraghava
@Ram0206
@cnha
@HeartcoRNews
@aakashd
@COEPs_BHAU
@GaneshNatarajan
@JCAE_JCCC
@audiolabrat
@IFrydman
@VanessaBoon
@kmore
@lhwoodruff
@sucoder
@DeniseDesu
You don’t have to choose one approach over the other if you use CogniSwitch. An innovative combination of proven techniques with guardrails based on facts and knowledge that makes #GenAI reliable @CogniSwitch @thequiet_one
A new graph-based RAG stack is emerging (GRAG?) 🔗Relation extraction (w/ models like REBEL) 🕸️Graph db: @neo4j, @NebulaGraph, @kuzudb, and more 🔎Querying: Text-to-cypher, or keyword extraction Here’s a guide 📕 on how you can build this new LLM stack in @llama_index 👇
Perfect prompts for a probabilistic engine without hallucinations - is that even a thing 🤔 Sounds like a perfect storm for a production system
People who say this don’t understand the difference between a lame demoware by noob AI engineers and a well-designed cracked team that developed incredible data enrichment and also wrote 10,000 evals that have perfect prompts for a production system that has zero hallucination
Apple not investing in OpenAI after all, new report says 9to5mac.com/2024/09/27/app… by @ChanceHMiller
Love this take on “wrapper” applications. The key is using the right layers in the #genai stack to provide the data and domain knowledge in the right form and structure @CogniSwitch @mtrajan @prasanna_says
A lot folks are negative on the application layer of ai. I keep hearing that the "LLM wrapper applications" will get disrupted by the LLM's. I think you could argue that back in 2006, when HubSpot was founded and SaaS was exploding that all of us were "database wrapper…
The proofs continue to pile up.
Can we really trust AI in critical areas like medical image diagnosis? No, and they are even worse than random. Our latest study, "Worse than Random? An Embarrassingly Simple Probing Evaluation of Large Multimodal Models in Medical VQA," uncovers the stark limitations of…
LLMs don't hallucinate in the company of @CogniSwitch
But LLMs can't do anything useful in the real world! They make a lot of mistakes! They hallucinate! They're so far away from utility! Blinky team - hold my beer... All shallow bugs going the LLM way hereon
And there are a number of them working on that problem in innovative ways that are very different from the current approaches @CogniSwitch
LLM - LVM for NL instructions RAG - LLM’s pipeline to external data Agents - WBS & task execution w/LLM
Tokens my dear Watson!
Our paper about reliably finding under-trained or 'glitch' tokens is out! We find up to thousands of these tokens in some #LLMs, and give examples for most popular models. arxiv.org/abs/2405.05417 More in 🧵
Interesting
Generating PowerPoints with Llama 3 🦙📊 This is an awesome article by @naivebaesian on how to use @llama_index to build a Llama3 RAG pipeline that doesn’t just answer questions, but can generate a Powerpoint slide deck. Python-pptx is a neat library to programmatically create…
Cool. A DB company with a copilot! The refreshing difference is that it uses a graph db with RAG. what enterprises have is a data problem especially with natural language data that encapsulates key SoPs and #knowledge that currently resides in digital docs. The richness and…
TigerGraph CoPilot addresses both the complex challenges associated with data management and analysis and the serious shortcomings of LLMs for business applications. By leveraging the power of natural language processing and advanced algorithms… CVSoci.al/y~egMC5F?cv=17…
I'll be attending TiEcon 2024: AI Ubiquity. Let me know if you're planning to attend so that we can say Hi! 👋 Or register now and join me at the event! whova.com/portal/registr… #TiEcon2024 #AIUbiquity #TiESV
I just wrote about why we are building @CogniSwitch link.medium.com/iC1gzcCUFIb
Going live with @anandprabhala in another 10 mins to talk about GenAI stack and how your engineering team should think about it. Join us here - linkedin.com/events/7181647…
Amen. As they say, it all starts with you. We need to share the collective intelligence of our teams with AI and then enjoy the levels to which reliable augmentation can take you
An incredible room...and, a wonderful exploration into the future of intelligent businesses and our role in shaping the next frontier. 🔭 My friend @dharmesh recently shared something that made me stop and think. "You compete with AI," he wrote. But it wasn't as much as a…
Wrote a short note on LI about getting #enterprise GenAI use cases into production linkedin.com/pulse/does-vec…
Amen. And look for other ways to do things it hallucinates on!
"Lord, grant me the insight to employ ChatGPT for the tasks it excels in, The patience to accept the tasks it cannot handle, And the wisdom to know the difference"
United States Trends
- 1. Ole Miss 30,9 B posts
- 2. Colorado 48 B posts
- 3. Travis Hunter 8.715 posts
- 4. Indiana 60,7 B posts
- 5. Kansas 24,6 B posts
- 6. Jaxson Dart 6.608 posts
- 7. Gators 18,2 B posts
- 8. Devin Neal 1.667 posts
- 9. Penn State 7.428 posts
- 10. Ohio State 39,8 B posts
- 11. Olivia Miles 1.376 posts
- 12. Wayne 136 B posts
- 13. Shedeur 6.078 posts
- 14. Lane Kiffin 4.682 posts
- 15. Heisman 7.019 posts
- 16. Kounde 23,3 B posts
- 17. Florida 134 B posts
- 18. Billy Napier 4.364 posts
- 19. Minnesota 16 B posts
- 20. Hannah Hidalgo 1.278 posts
Who to follow
-
Avinash Raghava 🇮🇳
@avinashraghava -
Clifton Road
@Ram0206 -
Ashish Sinha
@cnha -
HeartcoR Solutions
@HeartcoRNews -
aakash dharmadhikari
@aakashd -
BHAU@COEP
@COEPs_BHAU -
Ganesh Natarajan
@GaneshNatarajan -
Johnson County Adult Education
@JCAE_JCCC -
Skip Orvis
@audiolabrat -
Irving Frydman
@IFrydman -
Vanessa Boon (she/her)
@VanessaBoon -
Kerry Morrison
@kmore -
Laurie Woodruff
@lhwoodruff -
Prashant Parashar ()
@sucoder -
Desu
@DeniseDesu
Something went wrong.
Something went wrong.