7/ Self-Improving Code LLMs- generates pseudo data from knowledge gained through pre-training & fine-tuning; adds the data to the training dataset for the next step; shows that different code generation frameworks can be improved in performance.
AI
-
ChatGPT and GPT-4 Research Overview: Applications and Analysis
By
–
8/ Summary of ChatGPT/GPT-4 Research – an overview of applications of ChatGPT and GPT-4; the analysis is done on 194 relevant papers and discusses capabilities, limitations, concerns, and more.
-
Baize: Open-Source Chat Model Fine-Tuned with LoRA
By
–
5/ Baize – an open-source chat model fine-tuned with LoRA. Leverages 100K dialogs generated from ChatGPT chatting with itself; it releases the dialogs along with 7B, 13B, and 30B parameter models.
-
Machiavelli Benchmark: Evaluating LLM Ethics in Adventure Games
By
–
6/ Machiavelli Benchmark – a new benchmark of 134 text-based Choose-Your-Own-Adventure games to evaluate the capabilities and unethical behaviors of LLMs.
-
Eight Things to Know about LLMs: Capabilities and Limitations
By
–
3/ Eight Things to Know about LLMs – discusses important considerations regarding the capabilities and limitations of LLMs.
-
New 50-Page Survey on Large Language Models Released
By
–
4/ A Survey of LLMs – a new 50 pages survey on large language models.
-
Segment Anything Model releases billion mask segmentation dataset
By
–
1/ Segment Anything Model – a set of resources for image segmentation; releases the largest segmentation dataset with over 1B masks on 11M licensed images; the model’s zero-shot performance is competitive with or superior to fully supervised results.https://t.co/sJvKhmeECe
— DAIR.AI (@dair_ai) 9 avril 20231/ Segment Anything Model – a set of resources for image segmentation; releases the largest segmentation dataset with over 1B masks on 11M licensed images; the model’s zero-shot performance is competitive with or superior to fully supervised results.
-
GPT-4 Instruction Tuning Dataset for LLaMA Models
By
–
2/ Instruction Tuning with GPT-4 – a "first attempt" to use GPT-4 to generate instruction-following data for LLM fine-tuning; includes 52K unique English & Chinese instruction-following data used to instruction-tune LLaMA models.
-
Top ML Papers Week April 3-9 Segment Anything Model
By
–
Top ML Papers of the Week (April 3 – 9): – Segment Anything Model
– SegGPT
– A Survey of LLMs
– Instruction Tuning with GPT-4
– 8 Things to Know about LLMs
– Summary of ChatGPT/GPT-4 Research
… -
ChatGPT Plugin Extracts Wisdom from Lex Fridman Podcast
By
–
A ChatGPT plugin for getting pearls of wisdom from the @lexfridman podcast: https://t.co/n2QeFs9MAO
— Greg Brockman (@gdb) 9 avril 2023A ChatGPT plugin for getting pearls of wisdom from the @lexfridman podcast: