AI Briefing#llm
The latest AI and developer news about #llm, with the original source and a short summary.
The latest AI and developer news about #llm, with the original source and a short summary.
We benchmark the problem of handling high-cardinality categorical values that arises during LLM-based query analysis, and its solutions.
Healthify adopted OpenAI's API to revolutionize the accuracy and scalability of AI health coaching and food recognition technology.
ASUS and FuriosaAI are strengthening collaboration on next-generation chip and server design to address the AI inference market.
Dosu adopted the Evaluation Driven Development (EDD) approach to prevent performance degradation in LLM products and enhance reliability.
Paradigm adopted GPT-4 to increase the accuracy of clinical trial data extraction and dramatically improve patient accessibility.
This introduces how to use a text generation pipeline based on Optimum Habana to efficiently run Llama 2 models on Intel Gaudi 2 accelerators.
Mistral AI has launched its flagship model Mistral Large, featuring powerful reasoning and multilingual capabilities.
Hugging Face has released a guide on using PEFT and QLoRA to efficiently fine-tune Gemma models.
Google has released Gemma, an open LLM family built on Gemini technology.
LlamaIndex has unveiled LlamaCloud, a managed parsing, ingestion, and retrieval service.
LlamaIndex has unveiled LlamaCloud, a managed RAG pipeline featuring complex PDF parsing.
Hugging Face's PEFT library now supports new merging methods that can efficiently combine LoRA adapters.
This presents a method for building low-cost, high-efficiency custom models that achieve GPT-4-level performance by leveraging synthetic data generated by LLMs.
A major overhaul that separates the core and integration packages and deprecates ServiceContext.
AI21 and AWS demonstrated RAG- and TSM-based generative AI use cases at a hackathon.
The biggest obstacle to GenAI adoption was a lack of knowledge and infrastructure, and Clarivate built a prototype with AI21 in just two weeks.
Qwen1.5, the next-generation model in the Qwen series, has been released with various sizes and improved performance.
OpenAI announced new embedding models with improved performance and lower pricing, along with a GPT-3.5 Turbo update.
Qwen has released Qwen-VL-Plus and Qwen-VL-Max, multimodal models with greatly enhanced image reasoning and recognition capabilities.
RAG retrieves an organization's own documents to boost the accuracy and practicality of generative AI.
The **Qwen** series, aiming at AGI, has been released, including open-source **LLMs** and **LMMs** of various sizes.
Introduces the performance evaluation method and Model Merge technique of a model that achieved #1 on Hugging Face's Open LLM Leaderboard.
AI21 highlighted personalizing the shopping experience through LLM chatbots for retail and TSMs.
High-performance vector database **Qdrant** has launched a **Replit** template to help developers implement RAG and streamline onboarding.