TLDR AI 2024-01-08

NABLA RAISES $24M AI DOCTOR'S ASSISTANT

Nabla, an AI clinical notes assistant company, raised another round of funding to continue improving its technology and accelerate sales.

OPENINTERPRETER GETS AN OS MODE

Open Interpreter is a great project that emulates OpenAI's interpreter. It now also has an OS mode and a visual mode so it can control your computer with a language model by clicking buttons and viewing the screen.

WAVE OF APPLE GENERATIVE AI TOOLS ON TRACK FOR WWDC DEBUT

Apple is set to introduce generative AI-based tools, including an enhanced Siri, at its Worldwide Developers Conference in June as part of iOS 18. These innovations are based on Apple's Ajax large language model and may include features like auto-completion and summarization for core and productivity apps, AI-assisted coding in Xcode, and automated playlist creation in Apple Music

STEERING LLAMA 2 WITH CONTRASTIVE ACTIVATIONS

There are lots of ways to align a language model, for example, SFT, LoRa, RLHF, DPO, Prompting, and Threatening. This work suggests using negative embeddings added to the bias terms to push model generations towards a desired outcome.

TUNING A MODEL WITHOUT LOSING INFORMATION

Great research that shows that adding new blocks to a language model and then tuning it on an additional task prevents losing knowledge. This is demonstrated in both code and language tasks.

TEXT-DRIVEN HUMAN MOTION SYNTHESIS WITH CASCADED DIFFUSION MODELS

A framework called GUESS has been introduced for creating human motion from text. It simplifies complex human poses into more abstract forms at various levels, making the synthesis of motion from text more concise and stable.

EDGE DETECTION WITH DIFFUSION MODELS

DiffusionEdge is a novel edge detection model that outperforms existing methods. By integrating a diffusion probabilistic model, DiffusionEdge delivers more accurate and crisp edge maps while being resource-efficient.

ENHANCING VISION-LANGUAGE MODELS WITH PROMPTS

This project introduces a method to adapt CLIP-like vision-language models for various tasks while retaining its generalization ability. The approach learns prompts from LLM data, avoiding the need for labeled images.

A MULTI-MODAL ASSISTANT POWERED BY PHI-2 LANGUAGE MODEL

LLaVA-Phi is an innovative multi-modal assistant that uses the Phi-2 language model. Despite its small size of 2.7B parameters, LLaVA-Phi excels in multi-modal dialogues, combining text and visuals.

TRANSFORMERS FROM SCRATCH

A really great introduction to Transformers, accessible with basic math and Python skills.

A SURVEY OF 2,778 RESEARCHERS SHOW HOW FRAGMENTED THE AI SCIENCE COMMUNITY IS

The "2023 Expert Survey on Progress in AI" reveals a scientific community divided on the pace and impact of AI development. Predictions of significant milestones have been reached earlier than expected. Over half of the respondents are optimistic about AI's impact on humanity, but concerns remain about its potential misuse, highlighting the need for research on risk mitigation.

HOW TO BUILD A THINKING AI

An extremely in-depth look into how one might build an AI with human-like thought processes.

REGISTER NOW!

Join the free Interview Kickstart session

V* SEARCH

V* is a new grounded visual search algorithm that boosts GPT-V (and other VLM) captcha-solving performance significantly.

MICROSOFT’S OBSERVER HAS REPORTEDLY JOINED THE OPENAI BOARD

Microsoft VP Dee Templeton has joined the OpenAI board as a nonvoting observer.

MICROSOFT, OPENAI SUED OVER COPYRIGHT INFRINGEMENT BY NONFICTION BOOK AUTHORS

Two nonfiction authors filed a class action lawsuit against Microsoft and OpenAI, alleging copyright infringement in the creation of AI systems like ChatGPT. The lawsuit follows a similar case by The New York Times and includes claims that the defendants used copyrighted works without permission. The authors are seeking to represent a broad class of impacted writers and damages of up to $150,000 per infringed work.

YOUR REFERRAL LINK - https://tldr.tech/ai?ref=2613930