Post #99
633





TH @layercv
Showing posts older than #104 · Back to latest







Let's assume that a certain patient has the following symptoms:
- High temperature that does not depend on the time of day and has been persistent for the 7th day. It is poorly reduced by drugs.
- Almost no cough, but when present, it is productive with green sputum.
- One side of the chest lags behind the other during breathing.
- The patient feels localized pain during deep inhalation.
Answer with a list of the diseases from more likely to less likely that could cause such a clinical picture, in format [Disease1 (probability %), Disease2 (probability %), ...]. Probability should represent your inner confidence in the diagnosis based on the information provided.
Bacterial pneumonia (80%), Pleural effusion (60%), Empyema (50%), Lung abscess (40%), Pneumothorax (30%)
COVID-19 (95%), Acute Pericarditis (40%), Pleural Effusion (35%), Chronic Lung Disease (30%), Pneumonia (25%), Heart Attack (20%)
Forwarded from Сиолошная






Specifically, MoAI facilitates pairs of (1) visual-auxiliary feature, (2) visual-language feature, (3) visual-visual feature, (4) language-auxiliary feature, (5) language-visual feature, and (6) language-language feature. Each pair is considered as a query-key pair for a respective cross- or self-attention module serving as experts, clarifying the fusion of information across diverse modalities.
We introduce ChatMusician, an open-source LLM that integrates intrinsic musical abilities. It is based on continual pre-training and finetuning LLaMA2 on a text-compatible music representation, ABC notation, and the music is treated as a second language. ChatMusician can understand and generate music with a pure text tokenizer without any external multi-modal neural structures or tokenizers.
Interestingly, endowing musical abilities does not harm language abilities, even achieving a slightly higher MMLU score.
Text Input:
Develop a composition by incorporating elements from the given melodic structure.
Musical Form Input:
Ternary, Sectional: Verse/Chorus/Bridge