Links for 2023-05-21
Decision Theory with the Magic Parts Highlighted https://www.lesswrong.com/posts/5sFkZK342j5CmBCm8/decision-theory-with-the-magic-parts-highlighted (cf. what I wrote here: https://imgur.com/a/KTQoELy)
‘The last generation’: the young Chinese people vowing not to have children https://www.theguardian.com/world/2023/jan/20/the-last-generation-young-chinese-people-vow-not-to-have-children (Only 38% of Chinese college students want to EVER have children. 34% among women. Guessing at true desires of "uncertain" respondents, average Chinese college women desires just 0.94 kids. College men 1.05. https://www.sciencedirect.com/science/article/abs/pii/S1083318822002613)
"Researchers found that he had both amyloid plaques and tau tangles associated with Alzheimer’s disease. But crucially, tau was relatively limited in his entorhinal cortex, which is essential for memory." [Washington Post] https://archive.is/BgXOT
A temperate Earth-sized planet with tidal heating transiting an M6 star https://www.nature.com/articles/s41586-023-05934-8
Language Models Meet World Models: Embodied Experiences Enhance Language Models — “…our approach substantially improves base LMs on 18 downstream tasks by 64.28% on average.” https://arxiv.org/abs/2305.10626
Tree of Thoughts: Deliberate Problem Solving with Large Language Models — “…while GPT-4 with chain-of-thought prompting only solved 4% of tasks, our method achieved a success rate of 74%.” https://arxiv.org/abs/2305.10601
Universal Source Separation with Weakly Labelled Data: “The USS system can automatically detect and separate sound classes from a real recording. The USS system can separate up to hundreds of sound classes sound classes in a hierarchical ontology structure.” https://arxiv.org/abs/2305.07447
Optimizing Memory Mapping Using Deep Reinforcement Learning https://arxiv.org/abs/2305.07440
Augmenting Big LLMs with Small LLMs for Knowledge Guiding https://arxiv.org/abs/2305.04757
Small Models are Valuable Plug-ins for Large Language Models https://arxiv.org/abs/2305.08848
Large Language Models Can Be Used To Effectively Scale Spear Phishing Campaigns https://arxiv.org/abs/2305.06972
Dr. LLaMA: Improving Small Language Models in Domain-Specific QA via Generative Data Augmentation https://arxiv.org/abs/2305.07804