3/5 Enter semantic search! It uses text embeddings to turn words into vectors. It then uses similarity to find the most similar vector among responses to the query. The outcome? A more accurate, relevant response!
LLMS
-
Semantic Search: How Vector Embeddings Capture Meaning
By
–
4/5 Imagine this – the sentence "Hello, how are you?" & "Hi, what’s up?" will get similar vectors, while "Tomorrow is Friday" gets a different one. That's the power of semantic search!
-
Interest in LLM Applications for MLOps
By
–
I agree. Wonder too. It would be cool to see some works that apply LLMs in MLOps.
-
LLM Best Practices Meetups SF NYC Community Insights
By
–
I ran two LLM best practice meet-ups this month in SF and NYC. The energy, community and insights were awesome. If you missed it or didn’t have time to take notes (like me), @louicop has got you covered https://
medium.com/point-nine-new
s/learnings-exploring-the-gpt-llm-space-67a49d4e2e19
… -
Google Announces PaLM 2 Model Integration Across Products
By
–
Good thread by @DynamicWebPaige highlighting many of the new announcements about how we're using the PaLM 2 model throughout many Google products.
-
HuggingFace Launches Multi-Modal Agent API Chaining 100K Models
By
–
Excited to announce our multi-modal agent API that can automatically chain 100,000+ of HF models (stable diffusion, whisper, OpenAssistant,…) for text, audio, image, video, time-series,… based on your commands. All open-source so it can work locally! https://
huggingface.co/docs/transform
ers/transformers_agents
… -
Google Med-PaLM: Advanced Medical AI Language Model
By
–
Want to learn more about Med-PaLM? Read about it and watch the video at https://
sites.research.google/med-palm/ -
PaLM 2 Multiple Sizes Released Gecko Runs Locally
By
–
Waiting for more details on this but looks like there will be multiple sizes for PaLM 2. No param counts mentioned but that Gecko can run on a local device.
-
Meta AI’s Angela Fan Presents Text Generation Research in London
By
–
Welcoming @MetaAI 's Angela Fan at The Research and Applied AI Summit on June 23 in London! Angela focusses on research in text generation, including her recent work on No Language Left Behind and Universal Speech Translation for Unwritten Languages.
-
How AI Agents Work: Prompt Building and Chain-of-Thought
By
–
How does it work in practice? It's straightforward prompt-building:
• Tell the agent what it aims to do
• Give it tools
• Show examples
• Give it a task The agent uses chain-of-thought reasoning to identify its task and outputs Python code using the tools.