Good morning from the Times Center in New York City! The Abridge Keynote starts in about 15 minutes. On this thread, we will be sharing updates from inside the room.
Watch the livestream here: https://t.co/99YFgUrCX9
Evaluation of ambient scribes is a formidable task due to the free-form nature of generated text. Rigor requires automated metrics, strong benchmarks, clinician-in-loop trials, and in vivo testing. Learn how we're tackling these challenges at @AbridgeHQ:
https://t.co/8oWka8elwh
The D&I subsidy call for #EMNLP2024 is now available: https://t.co/0II2uKwebk
Subsidies available for registration, caregiving, bandwidth, travel and VPN! See website for more details.
Deadline is Friday, October 11th 11:59 (AOE)
NSF and @SimonsFdn are launching two new National AI Research Institutes. Led by @UTAustin and @NorthwesternU, each institute will receive $20 million to accelerate humanity’s understanding of the universe through the development of new AI-powered tools. https://t.co/3uW5tV5r92
new post - we took a look at behind the technical curtain some of the interesting engineering challenges behind a company (@AbridgeHQ ) training their own LLMs.
-dealing with multiple languages
-handling model drift
-generalist models vs. healthcare specific ones
and more
(just fyi, this one is a sponsored post but it's pretty interesting if you're interested in healthcare-specific LLMs)
Two of our brilliant team members, Davis Liang and Elisa Ferracane, represented Abridge at ACL 2024 in Bangkok, showcasing research that pushes the boundaries of Natural Language Processing (NLP) and multilingual understanding. 🌎🌟
Davis Liang, a Staff Research Scientist at Abridge, presented Belebele, a multilingual reading comprehension dataset covering 122 language variants designed to tackle the global challenge of understanding text across diverse languages. This dataset enables direct comparison of how well models understand different languages and opens new possibilities for evaluating and discussing LLM multilinguality: https://t.co/xeqgJnmwWo
Elisa Ferracane, a Senior NLP Research Scientist at Abridge, in collaboration with lead author Sanjana Ramprasad, an Abridge research intern, shared cutting-edge research on tackling hallucination errors in LLMs, particularly in dialogue summarization. Their work introduced a new error category to better address these challenges, marking a crucial step forward as LLMs evolve: https://t.co/51ONRrVBGk
We’re proud to have these pioneers driving innovation at Abridge, where their expertise guides us as we build the most advanced AI platform for clinical documentation. Congrats, Davis, Elisa, and Sanjana for all your outstanding work! 👏
Headed to ACL? Let’s talk about factuality and summarization, especially in the domain of conversations. DM me here or on Whova.
You can also find me at the Tuesday morning poster session, presenting our paper with @sanjana_rampi.
https://t.co/dqnYnI1VUV
#ACL2024NLP
On my way to @naaclmeeting 🇲🇽 to present “TOPICAL: TOPIC Pages AutomagicaLly 🪄📄”
In this paper, we ask the simple question: can LLMs be used to write high-quality scientific topic pages automatically? 🧵👇
Paper: https://t.co/LLFI6cDCIk
Demo: https://t.co/8aW2LkvgA8
[1/7]
Focusing on one of my favorite domains of conversations, where so much is left unsaid, @sanjana_rampi shares lots of insights on understanding and identifying hallucinations in summaries:
Check out our #ACL2024NLP paper benchmarking LLMs for faithfulness on dialogue summarization.
Our analysis highlights hallucination subtleties and introduces a new error category to capture LLM-specific behavior 🧵👇
https://t.co/VyvnlnhwFM
w/ @ElisaFerracane@zacharylipton
@ReviewAcl We have an issue with the author list for our ARR submission which is now accepted to ACL 2024.
Can you please help? (We've tried emailing both arr and acl mailing lists with no response).
@ReviewAcl Please allow for smaller max load in the review form.
I am a careful reviewer and unable to commit to reviewing 4 papers.
I feel this policy targets and excludes researchers who are already being marginalized due to lack of resources.
@thamar_solorio@annargrs
📣 Today, we are announcing our $150M Series C! This is one of the largest funding rounds made to date in generative AI for healthcare.
We are grateful to the luminary investors and iconic institutions who believe in us and our mission—to power deeper understanding in healthcare. This round was led by @lightspeedvp who will be joining our board. Other new and existing investors joining the round include co-lead @Redpoint, with support from @IVP, @sparkcapital, @usv, @BessemerVP, @WittingtonVC, Mass General Brigham Artificial Intelligence and Digital Innovation Fund (AIDIF), Kaiser Permanente Ventures, and CVS Health Ventures.
With this new capital, we will continue to push boldly into fundamental research, developing bedrock foundation models that draw upon vast troves of multimodal healthcare data.
Looking forward to this new wave of possibilities as we continue to fulfill our vision of improving the lives of patients and clinicians.
Read more about our Series C here: https://t.co/cYC4Dt6iae
We are kicking-off November with an Open Session on Multilingual Language Modeling. @LiangDavis will join us for our Thursdays with Science – Open Session.
The session will take place on the 2nd of Nov., at 2pm (WEST). Register at https://t.co/z0Kc01wJpW and join us!
🎉 Abridge is proud to announce a Series B raise of $30 million, led by AI luminary @sparkcapital. The round includes existing investor @BessemerVP along with several leading healthcare innovators (read more below).
Attending the Machine Learning for Healthcare conference in New York this Friday and Saturday? Stop by our booth for a real-time demo and learn how we’re improving the care experience for clinicians and their patients with thoughtful AI. https://t.co/XeH6LNMhEZ
Our engineering teams are growing! If you're an experienced mobile engineer looking for a fast-paced environment and excited by the prospect of unlocking the potential of AI in healthcare, apply to join us at @AbridgeHQ.
https://t.co/FOFBfZQx7f
https://t.co/BfrujJqLU7): interview this AM on @cbschicago about ChatGPT and UT Austin's own @UTGoodSystems grand challenge on the design, development, and evaluation of responsible AI technologies. @UTAustin@UTiSchool@EngagingNews