301
Arc Prize 2024 Winners and Technical Report
Official winners and technical report.
Topics /
AI & Machine Learning
The most popular blog posts about ai & machine learning on Hacker News. 363 posts.
302
Pgvector Is Now Faster Than Pinecone at 75% Less Cost
Meet pgvectorscale, an open-source PostgreSQL extension that supercharges pgvector’s vector search performance for large-scale vector workloads.
303
LLM-based sentiment analysis of Hacker News posts between Jan 2020 and June 2023
We analyzed topics covered by 100k Hacker News posts and the sentiments they evoke using Metaflow.
304
305
306
Mining JIT traces for missing optimizations with Z3
In my last post I've described how to use Z3 to find simple local peephole optimization patterns for the integer operations in PyPy's JIT. An example is int_and(x, 0) -> 0. In this post I want to scal
307
308
Llama 3 feels significantly less censored than its predecessor
Compared to Llama 2, Llama 3 feels much less censored. Meta has substantially lowered false refusal rates. Llama 3 will refuse less than 1/3 of the prompts previously refused by Llama 2.
309
MK1 Flywheel Unlocks the Full Potential of AMD Instinct for LLM Inference
AMD acquires MK1, welcoming their world-class team and marking another key milestone in our strategy to advance AI performance and efficiency.
310
LLM-generated code must not be committed without prior written approval by core
If you are not sure if the code you plan to commit is acceptable (e. g. when taking code that was submitted with a problem report), ask a developer who is familiar with that part of the system for review. If you're new to the Project, check with your sponsor.
311
PaliGemma: Open-Source Multimodal Model by Google
Learn how to use PaliGemma, a vision language model developed and released by Google that has multimodal capabilities.
312
Speaking without vocal cords, thanks to a new AI-assisted wearable device
The adhesive neck patch is the latest advance by UCLA bioengineers in speech technology for people with disabilities.
313
Learning From DNA: a grand challenge in biology
Is DNA all you need?
314
Building an early warning system for LLM-aided biological threat creation
We’re developing a blueprint for evaluating the risk that a large language model (LLM) could aid someone in creating a biological threat. In an evaluation involving both biology experts and students, we found that GPT-4 provides at most a mild uplift in biological threat creation accuracy. While this uplift is not…
315
316
Machine Learning Is Still Too Hard for Software Engineers
Machine Learning has made huge strides in the last few decades, but it remains largely inaccessible to the average developer. This post examines some of the reasons why.
317
318
Extending the context length to 1M tokens
API Documentation (Chinese) HuggingFace Demo ModelScope Demo Introduction After the release of Qwen2.5, we heard the community’s demand for processing longer contexts. In recent months, we have made many optimizations for the model capabilities and inference performance of extremely long context. Today, we are proud…
319
Linear transformers are faster after all
Our mission is to train a neural network to model all human output.
320
Mistral AI raises $640M at $6B valuation
Today, we are excited to triple down on our partnership with Mistral AI by co-leading their €600M Series B round.
321
Qwen1.5-110B
GITHUB HUGGING FACE MODELSCOPE DEMO DISCORD Introduction Recently we have witnessed a burst of large-scale models with over 100 billion parameters in the opensource community. These models have demonstrated remarkable performance in both benchmark evaluation and chatbot arena. Today, we release the first 100B+ model…
322
A new artificial intelligence tool for cancer
ChatGPT-like AI model can diagnose, guide treatment, predict survival for multiple cancer types
323
Models all the way down
LAION-5B is an open-source foundation dataset. It contains 5.8 billion image and text pairs—a size too large to make sense of. We follow the construction of the dataset to better understand its contents, implications and entanglements.
324
Anthropic: Expanding Access to Claude for Government
Claude 3 Haiku and Sonnet are now available in the AWS Marketplace for the US Intelligence Community and in AWS GovCloud.
325
Viking 7B: open LLM for the Nordic languages trained on AMD GPUs
Silo AI is announcing the release of the first open LLM for the Nordic languages
326
Using the term 'AI' in product descriptions reduces purchase intentions
Companies may unintentionally hurt their sales by including the words “artificial intelligence” when describing their offerings that use the technology, according to a study led by WSU researchers.
327
INTELLECT–1: Launching the First Decentralized Training of a 10B Parameter Model
We're excited to launch INTELLECT-1, the first globally-distributed training run of a 10-billion-parameter model, inviting anyone to contribute compute and participate. This brings us one step closer towards open source AGI.
328
Turning the Tables on AI
How about using it not to think less but *more*?
329
Leveraging AI for efficient incident response
We’re sharing how we streamline system reliability investigations using a new AI-assisted root cause analysis system. The system uses a combination of heuristic-based retrieval and large language m…
330
Inversion: Fast, Reliable Structured LLMs
We create the world's fastest and most efficient AI.