common themes: AGI
agents
much better 4o upgrade
much better memory
longer context
“grown up mode”
deep research feature
better sora
more personalization (interestingly, many great updates we have coming were mentioned not at all or very little!)
MULTIMODAL AI
-
OpenAI Upcoming Features: AGI, Agents, and Advanced Capabilities
By
–
-
Vector Store and Matching Engine: Fast Similarity Search with Deep Learning
By
–
Vector Store and Matching Engine performs searches on large stores of vector embeddings from text, images, documents & more in milliseconds using #DeepLearning models: https://
abacus.ai/vectorstoreVec
tor
… Similarity Search is QBE (Query-By-Example). This is MUCH BETTER… @abacusai -
Fine-tuning video models: OpenAI behind?
By
–
Fine-tuning video models will be big. I'm doubtful that OpenAI will lead the pack on this, they seem to be behind with video/image gen.
-
Top 25 AI Models 2024: Flux, Llama, Gemma, Grok
By
–
Top 25 AI models in 2024 on Hugging Face @bfl_ml w/ Flux.1-dev & Flux.1-schnell – current SoTA open Text to Image models @AIatMeta w/ Llama 3.X series (1B to 70B) – competitive LLMs across sizes @StabilityAI w/ SD 3.5 Medium & Large @GoogleAI Gemma 7B @xai w/ grok-1
-
Cross-Modal Video VAE for Large Motion Video Autoencoding
By
–
Large Motion Autoencoding with Cross-modal VAE https://
arxiv.org/abs/2412.17805 -
AgiBot World: Large-Scale Robotic Learning Dataset Launch
By
–
Towards the "ImageNet Moment" for Embodied AI.Introducing AgiBot World, the first large-scale robotic learning dataset designed to advance multi-purpose robotic policies. HuggingFace：
https://
huggingface.co/agibot-world
Github：
https://
github.com/OpenDriveLab/a
gibot-world
… https://
agibot-world.com -
Adversarial Examples for Vision-Language Models via Diffusion
By
–
Efficient Generation of Targeted and Transferable Adversarial Examples for Vision-Language Models Via Diffusion Models https://
arxiv.org/abs/2404.10335 -
Grok and InVideo Create Beautiful Science Fiction Content
By
–
OK Elon here is some positive, beautiful and informative content. A SciFi video written by Grok and visualized by InVideo: https://t.co/nESPufbmrG
— Bob Gourley – e/acc (@bobgourley) 29 décembre 2024OK Elon here is some positive, beautiful and informative content. A SciFi video written by Grok and visualized by InVideo:
-
Question about reading letters in indigenous languages
By
–
Oh, did someone submit a letter in an indigenous language? How did you read it?
-
CoMCTS: Enhancing Multimodal Models with Step-by-Step Reasoning
By
–
8). Empowering MLLM with o1-like Reasoning and Reflection – proposes a new learning-to-reason method called CoMCTS that enables multimodal language models to develop step-by-step reasoning capabilities by leveraging collective knowledge from multiple models.