From $P(y|x)$ to $P(y)$: Investigating Reinforcement Learning in Pre-train Space
Paper โข 2604.14142 โข Published โข 28
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
ReMix: Reinforcement routing for mixtures of LoRAs in LLM finetuning
Towards a Neural Debugger for Python
Benjamin-eecsย
authored a
paper about 11 hours ago
Benjamin-eecsย
authored a
paper 30 days ago
q-rzย
submitted a
paper to Daily Papers about 1 month ago
kaiyuyueย
submitted a
paper to Daily Papers about 2 months ago
taesiriย
submitted a
paper to Daily Papers 2 months ago
taesiriย
submitted 2
papers to Daily Papers 3 months ago
lgmetaย
submitted a
paper to Daily Papers 3 months ago
taesiriย
submitted 2
papers to Daily Papers 3 months ago
taesiriย
submitted a
paper to Daily Papers 4 months ago
Benjamin-eecsย
authored 2
papers 6 months ago
SelmaNajih001ย
posted an update 6 months ago
Post
2944
How Financial News Can Be Used to Train Good Financial Models ๐ฐ
Numbers tell you what happened, but news tells you why.
Iโve written an article explaining how news can be used to train AI models for sentiment analysis and better forecasting. Hope you find it interesting!
Read it here: https://huggingface.co/blog/SelmaNajih001/llms-applied-to-finance
I would love to read your opinions! Iโm open to suggestions on how to improve the methodology and the training
Numbers tell you what happened, but news tells you why.
Iโve written an article explaining how news can be used to train AI models for sentiment analysis and better forecasting. Hope you find it interesting!
Read it here: https://huggingface.co/blog/SelmaNajih001/llms-applied-to-finance
I would love to read your opinions! Iโm open to suggestions on how to improve the methodology and the training
- 1 reply
SelmaNajih001ย
posted an update 6 months ago
Post
3055
Which is the best model to use as a signal for investment?
Here who is gaining the most:
https://huggingface.co/spaces/SelmaNajih001/InvestmentStrategyBasedOnSentiment
The Space uses titles from this dataset:
๐ SelmaNajih001/Cnbc_MultiCompany
Given a news title, it calculates a sentiment score : if the score crosses a certain threshold, the strategy decides to buy or sell.
Each trade lasts one day, and the strategy then computes the daily return.
For Tesla the best model seems to be the regression ๐
Just a quick note: the model uses the closing price as the buy price, meaning it already reflects the impact of the news.
Here who is gaining the most:
https://huggingface.co/spaces/SelmaNajih001/InvestmentStrategyBasedOnSentiment
The Space uses titles from this dataset:
๐ SelmaNajih001/Cnbc_MultiCompany
Given a news title, it calculates a sentiment score : if the score crosses a certain threshold, the strategy decides to buy or sell.
Each trade lasts one day, and the strategy then computes the daily return.
For Tesla the best model seems to be the regression ๐
Just a quick note: the model uses the closing price as the buy price, meaning it already reflects the impact of the news.
Benjamin-eecsย
authored a
paper 6 months ago
SelmaNajih001ย
posted an update 6 months ago
Post
393
Which is the best model to use as a signal for investment? ๐ค
Iโve created a Space where you can compare three models:
-Two available on my profile
- ProsusAI/finbert
You can try it here:
๐ https://huggingface.co/spaces/SelmaNajih001/InvestmentStrategyBasedOnSentiment
The Space uses titles from this dataset:
๐ SelmaNajih001/Cnbc_MultiCompany
Given a news title, it calculates a sentiment score : if the score crosses a certain threshold, the strategy decides to buy or sell.
Each trade lasts one day, and the strategy then computes the daily return.
Just a quick note: the model uses the closing price as the buy price, meaning it already reflects the impact of the news.
If I had chosen the opening price, the results would have been less biased but less realistic given the data available.
Iโve created a Space where you can compare three models:
-Two available on my profile
- ProsusAI/finbert
You can try it here:
๐ https://huggingface.co/spaces/SelmaNajih001/InvestmentStrategyBasedOnSentiment
The Space uses titles from this dataset:
๐ SelmaNajih001/Cnbc_MultiCompany
Given a news title, it calculates a sentiment score : if the score crosses a certain threshold, the strategy decides to buy or sell.
Each trade lasts one day, and the strategy then computes the daily return.
Just a quick note: the model uses the closing price as the buy price, meaning it already reflects the impact of the news.
If I had chosen the opening price, the results would have been less biased but less realistic given the data available.
Leitianย
authored a
paper 6 months ago
Benjamin-eecsย
authored a
paper 6 months ago
Leitianย
authored a
paper 6 months ago