These new phi-1 & 1.5 models are fascinating, no? performances crushing 10 times bigger models secret sauce coming from a magic textbook dataset close to zero information on this dataset other than GPT3.5 generated Is it time for an "Open Textbook" project??
LLMS
-
Evolutionary Algorithms Optimize Large Language Model Prompts
By
–
"Connecting Large Language Models with Evolutionary Algorithms Yields Powerful Prompt Optimizers" What's going on here:
A new method is proposed to automatically generate good prompts for large language models by combining evolutionary algorithms with the models. What does this -
Linear Autoregressive Predictors Learn All Turing-Computable Functions
By
–
“Any Turing-computable function can be learned with a linear autoregressive next token predictor.”
-
How Often Do LLMs Think About the Roman Empire?
By
–
How often does your LLM ‘think’ about the Roman Empire?
-
RLHF and LLM Evaluations with Nathan Lambert
By
–
Generating Conversation: RLHF and LLM Evaluations with Nathan Lambert (Episode 6) https://
bit.ly/3Rlp1Hd
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
LLMs Water Consumption Environmental Impact Sustainability Concerns
By
–
Does this include the fact LLMs have been shown to use 1/2 a litre of water per 15 minute usage / query? Said respectfully but AI/ML by and large is not helping as much as hurting water and other issues.
-
MAmmoTH: Open-Source LLMs for Mathematical Problem-Solving
By
–
10/ MAmmoTH – a series of open-source LLMs tailored for general math problem-solving; the models are trained on a curated instruction tuning dataset and outperform existing open-source models on several mathematical reasoning datasets.
-
Radiology-Llama2: Specialized LLM for Medical Imaging
By
–
8/ Radiology-Llama2: Best-in-Class LLM for Radiology – presents an LLM based on Llama 2 tailored for radiology; it's tuned on a large dataset of radiology reports to generate coherent and clinically useful impressions from radiology findings.
-
LLMs Self-Align Without Finetuning via Self-Boosting
By
–
4/ LLMs Can Align Themselves without Finetuning? – discovers that by integrating self-evaluation and rewind mechanisms, unaligned LLMs can directly produce responses consistent with human preferences via self-boosting.
-
Survey of Hallucination Phenomena in Large Language Models
By
–
6/ A Survey of Hallucination in LLMs – classifies different types of hallucination phenomena and provides evaluation criteria for assessing hallucination along with mitigation strategies.