Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models Chen et al.: https://
arxiv.org/abs/2401.01335 #Artificialintelligence #DeepLearning #MachineLearning
RESEARCH
-
Self-Play Fine-Tuning Converts Weak Language Models Strong
By
–
-
AI Revolutionizes Urban Planning and Smart City Simulation
By
–
Shaping Future Cities: The Revolutionary Role of AI in Urban Planning and Simulation https://
linkedin.com/pulse/shaping-
future-cities-revolutionary-role-ai-urban-planning-babin-s34je
…
#ArtificialInteligence #urbanplanning #SmartCities #innovation #technology @jblefevre60 @kalydeoo @Ym78200 @ipfconline1 @LaurentAlaus @Fabriziobustama @enilev @3itcom -
Essential Steps for Successful AI Projects: A Complete Guide
By
–
Great article unpacking the steps for #AI projects. Crucial insights for #AIConsultants! Understand the Problem Data Collection & Prep Model Building Testing & Validation Deployment Monitoring & Maintenance
-
SOLAR-10.7B Model Enhanced with Direct Preference Optimization
By
–
#solarllm SOLAR-10.7B + Direct Preference Optimization (DPO)
-
Self-Extend LLM: Extending Context Window Without Fine-tuning
By
–
#solarllm #16K Very interesting work, Self-Extend LLM! "The proposed method does not require any training. With only four lines of code modification, it can effortlessly extend existing LLMs’ context window without any fine-tuning. We conducted comprehensive experiments, and the
-
AI Predictions for 2024: Expert Forecasts and Timeline Analysis
By
–
I busted out my crystal ball for this one. Here're my predictions of what will happen in AI in 2024. I'm probably wrong about a lot of them (especially the timelines) but it was still fun to make and share my thoughts. Share your predictions as well! https://
youtu.be/UEsSL3Wdcfc -
Precision Engineering: Key to ML Progress and System Understanding
By
–
One trick to making ML progress is repeatedly increasing the precision of the engineering and decreasing the number of mysteries being papered over — i.e. just focusing on improving your understanding of the system you're working with.
-
Contrastive Loss with Reconstruction in Two-Step Training
By
–
I think you have to use contrastive loss still on top of the reconstruction, in two steps — r u doing that?
-
Diffusion Model Training Optimizations and Telephoning Data Augmentation
By
–
if you want to read the paper it's here; they developed a bunch of cool optimizations for diffusion model training too, and a cool data augmentation technique called 'telephoning'… http://
arxiv.org/abs/2310.16825 -
Early AI Image Generation Outputs: When Broken Looks Like Progress
By
–
i remember i was on campus working on my own stuff, it was 2am the night before a paper deadline, and he shows me these super cursed pictures of a pixar character and goes "that's elsa from frozen!! it's working!!!!" never seen someone so excited to see outputs that look broken