Unreal… @alexalbert__ please tell me this is emergent and you didn’t train it in
RESEARCH
-
LLMs Demonstrate Creative Capabilities Across Novel Complex Tasks
By
–
I would add that by every measure of creativity we have (all of which are flawed, which didn't matter when we were only measuring humans) LLMs produce creative results And frontier LLMs operate on a high level on novel managerial, medical, and legal tasks. Its all kind of weird.
-
Batch Shuffling Causes Accuracy Regression to Baseline
By
–
shuffled each batch and accuracy regressed down near the baseline
-
Contrastive Learning Bug Discovery in AGI Development
By
–
ok before one of you tries to assassinate me for building AGI, i figured out the bug. at least it's an interesting one so we're doing contrastive learning, which is a matching task between (query, document) pairs query 1 matches to document 1, query 2 matches to document
-
Snorkel AI: NLP Engineer Shares Year-Long Implementation Insights
By
–
James Dunham, an NLP Engineer at Georgetown's CSET, offers key insights from a year of using Snorkel AI. He shared the experiences of several project teams, discussing the challenges they faced, how Snorkel helped, and the obstacles that remained.
-
Argonne Hosts AI Accelerator Training Workshop Cerebras
By
–
Attention all Open Science Contributors! @argonne is hosting a two-day training workshop at the @argonne_lcf AI Testbed that includes Cerebras AI accelerators. Registration ends on April 5th. This workshop is designed for universities, national laboratories, and industry
-
DSPy and LangChain LLM Systems Optimization Webinar
By
–
Webinar: Optimization of LLM Systems with DSPy and LangChain/LangSmith Sign up here: https://
us06web.zoom.us/webinar/regist
er/WN_n4HUNqhFQFa4ax3kYrZW8g#/registration
… There are a lot of novel and interesting ideas in DSPy around optimization of LLM systems Super excited to host a webinar with Omar K. and
@hwchase17 diving into -
Google VLOGGER AI generates video avatars from images
By
–
Google's VLOGGER AI model can generate video avatars from images – what could go wrong? This AI technology could lead to more relatable helpdesk agents or a new wave of convincing deepfakes. https://
zdnet.com/article/google
s-vlogger-ai-model-can-generate-video-avatars-from-images-what-could-go-wrong/
… @GoogleAI -
AI Understands Console Formatting Without Library Installation
By
–
The most mind blowing thing here to me is that it understands the console formatting output without having the library installed to do that
-
Accidental Machine Language: Why Academics Lack Urgency
By
–
I am pretty surprised by the lack of urgency among many (but not all) academics in addressing what seems to be one of the biggest developments in modern times – we accidentally built a machine that produces something that looks like language & thought Why? What does it teach us?