Self-supervised learning and transformer-based technologies in breast cancer imaging - frontiersin.org
<a href="https://news.google.com/rss/articles/CBMikAFBVV95cUxQMXpqMzhlSDlwWDVGNGpEd19BcG8yTHN3OF8wZmktT3pzY1l0RjN2UHd0QVR3akFNWkI4S0ZuQ1l6QVdydlVrZHJRTWdNemFFdzhuMG42ZG5XUGtuU0laR0w3bVJHM255alBNVDU1MUllMUdNU3IxU2w2UV82WE1YYUtFWnpCeHRGYW9TdGJqTVA?oc=5" target="_blank">Self-supervised learning and transformer-based technologies in breast cancer imaging</a> <font color="#6f6f6f">frontiersin.org</font>
Could not retrieve the full article text.
Read on GNews AI transformer →Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
transformer
Google - Gemma 4 now in Unsloth!
Google releases Gemma 4 with four new models: E2B, E4B, 26B-A4B, 31B. You can now run and train the Gemma 4 models in Unsloth. Guide / Blog: https://unsloth.ai/docs/models/gemma-4 The multimodal reasoning models are licensed under Apache 2.0. Run E2B and E4B on 6GB RAM, and on phones. Run 26B-A4B and 31B on ~18GB. GGUFs: https://huggingface.co/collections/unsloth/gemma-4 Updates Tool calls for smaller models are now more stable and don't cut off anymore Context length is now properly applied. Tool calls for all models are now +30% to +80% more accurate. Web search now actually gets web content and not just summaries Number of tool calls allowed are increased to 25 from 10 Tool calls now terminate much better, so looping / repetitions will be reduced More tool call healing and de-duplicatio
700KB embedding model that actually works, built a full family of static models from 0.7MB to 125MB
Hey everyone, Yesterday I shared some static embedding models I'd been working on using model2vec + tokenlearn. Since then I've been grinding on improvements and ended up with something I think is pretty cool, a full family of models ranging from 125MB down to 700KB, all drop-in compatible with model2vec and sentence-transformers. The lineup: Model Avg (25 tasks MTEB) Size Speed (CPU) potion-mxbai-2m-512d 72.13 ~125MB ~16K sent/s potion-mxbai-256d-v2 70.98 7.5MB ~15K sent/s potion-mxbai-128d-v2 69.83 3.9MB ~18K sent/s potion-mxbai-micro 68.12 0.7MB ~18K sent/s Evaluated on 25 tasks (10 STS, 12 Classification, 3 PairClassification), English subsets only. Note: sent/s is sentences/second on my i7-9750H These are NOT transformers! they're pure lookup tables. No neural network forward pass at
Meta-Programming and Macro capabilities of various languages
Meta-programming = the broad idea of “programs that manipulate or generate programs” . It can happen at runtime (reflection) or compile-time (macros). Macros = one specific style of meta-programming, usually tied to transforming syntax at compile time (in a pre-processor or AST-transformer). It takes a piece of code as input and replaces it with another piece of code as output, often based on patterns or parameters. Rule‑based transformation: A macro is specified as a pattern (e.g., a template, an AST pattern, or token pattern) plus a replacement that is generated when that pattern is matched. Expansion, not function call: Macro use is not a runtime call; the macro is expanded before execution, so the final code is the result of replacing the macro invocation with its generated code. Here
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!