Skip to main content

Ctrl+K

LLMs <3 MIR

I. Large Language Models

1. Basics: Before the LLMs
2. Architectures
- 2.1. Transformer
- 2.2. Models
3. Tokenizers
- 3.1. Word as a Token (Pre-LLM era)
- 3.2. Sub-word Tokenizers
4. Pre-Training, and the Induced Properties
- 4.1. Next-Token Prediction
- 4.2. A LLM is What It Reads
5. Finetuning
- 5.1. Instruction Finetuning
- 5.2. What LLMs Learn during Finetuning
6. Evaluation

II. LLM as a Tool with Common Sense

1. LLMs are Zero-shot Learners
2. Dataset Cleaning and Creation
3. Music Information Extraction
4. Music Recommendation
5. Music Composition Assistance

III. Multimodal LLMs

1. Multimodal LLMs
2. Multimodal Input Processing
3. Music Audio LLMs

IV. Weakness of LLMs for MIR

1. Weaknesses of LLMs for MIR
2. LLMs can’t rhyme
3. LLMs don’t know every artist

V. Finale

1. Finale

Resources

References

Repository
Open issue

.md

Finetuning

5. Finetuning#

previous

4.2. A LLM is What It Reads

next

5.1. Instruction Finetuning

By Keunwoo Choi

© Copyright 2023.