How come LLMs have read AI textbooks and still aren't able to reason?
LLMS
-
Practical daily AI workflow using custom GPTs and prompt engineering
By
–
Mostly prompting ad-hock or use my own custom GPTs. “record what I say and make it better readable” is my daily writing prompt to convert braindump into news posts. Custom GPTs for generating news titles based on Google News guidelines – humanizing text to rephrase a list of
-
Comparative analysis of generative AI image models
By
–
More results out of my standard testing prompts "cyberpunk hacker robot". In general, it is quite comparable to Flux and it is also able to generate text very well
-
SRAM Optimization Accelerates Transformer Self-Attention Performance
By
–
"When implementations of the Transformer's self-attention layer utilize SRAM instead of DRAM, they can achieve significant speedups." Thanks @moritzthuening
, read the full paper here –> -
Chat History Storage: Alternating vs Single Message Architecture
By
–
Slightly off-topic, but I'm curious if you've compared storing the chat history as a sequence of alternating User and Assistant Messages (the natural/expected form) vs just putting everything into a single User Message so the prompt is always just 1 System and 1 User Message.
-
Grok-2: Elon Musk’s Latest AI Model Capabilities
By
–
Elon Musk’s Grok-2: Bold or Reckless? Dive into the wild world of Elon Musk's latest AI creation—Grok-2! This generative AI model from xAI, available through X Premium, is making waves by matching capabilities with GPT-4 in coding and math. But Grok-2 is not just
-
Salesforce Introduces xGen-MM: Open Large Multimodal Models Framework
By
–
Salesforce presents xGen-MM (BLIP-3) A Family of Open Large Multimodal Models discuss: https://
huggingface.co/papers/2408.08
872
… This report introduces xGen-MM (also known as BLIP-3), a framework for developing Large Multimodal Models (LMMs). The framework comprises meticulously curated -
Automated Design of Agentic Systems Using Foundation Models
By
–
Automated Design of Agentic Systems discuss: https://
huggingface.co/papers/2408.08
435
… Researchers are investing substantial effort in developing powerful general-purpose agents, wherein Foundation Models are used as modules within agentic systems (e.g. Chain-of-Thought, Self-Reflection, -
JPEG-LM: Using LLMs as Image Generators with Canonical Codecs
By
–
JPEG-LM LLMs as Image Generators with Canonical Codec Representations discuss: https://
huggingface.co/papers/2408.08
459
… Recent work in image and video generation has been adopting the autoregressive LLM architecture due to its generality and potentially easy integration into multi-modal -
Context sharpens AI prompts with specifics, expectations, audience
By
–
Context sharpens AI prompts. Essentials for precision and relevance. • Include specific details
• Set clear expectations
• Tailor for audience needs Read more: https://
buff.ly/3MbCIVm