Unveiling the AI Potential: FinBert beats ChatGPT in Financial Text … – Analytics India Magazine
There are several aspects in a study conducted by JPMorgan and Queen’s University where FinBert, an LLM fine-tuned on financial domain data, specific terminology, language structures, and concepts demonstrated its superiority over ChatGPT in the context of financial text analytics.
Gpt-3.5-turbo and GPT-4 with 8k tokens were compared with FinBert, and tested on Arithmetic Reasoning, News Classification Sentiment Analysis and Named Entity Recognition.
FinBert outperforms ChatGPT in sentiment analysis tasks related to financial texts. Sentiment analysis in finance requires understanding nuanced expressions and the impact of news on investors.
FinBert, being specifically designed for the financial domain, does not require extensive adaptation or fine-tuning to perform well in these tasks. In Few-shot Learning: ChatGPT’s performance on various tasks required more extensive prompts.
FinBert outperforms ChatGPT and even competes with human experts in arithmetic reasoning. This indicates that FinBert is a highly specialised model for financial tasks, while ChatGPT, might not reach the level of expertise demonstrated by FinBert in the financial domain.
In tasks such as financial named entity recognition (NER) and sentiment analysis, where a deep well of domain-specific knowledge is essential, ChatGPT and GPT-4 struggles. Their inability to grasp the intricacies of financial terminologies becomes evident.
Comparative analysis puts both the models against fine-tuned models tailored for the financial sector, like FinBert and FinQANet. The outcome underscores the fact that, while these LLMs hold potential, they are not yet on par with their specialised counterparts.
This study paves the way for further enhancements.
The gap between these state-of-the-art generative language models and domain-specific proficiency remains, but it also presents a promising opportunity for refinement.
Rajiv Shah, a machine learning engineer at Hugging Face said on linkedin, “A domain-specific model like FinBERT is more accurate for finance tasks than GPT-4”.
The most prestigious AI awards in the country. Nominations Open.
Discover special offers, top stories, upcoming events, and more.
Stay Connected with a larger ecosystem of data science and ML Professionals
To use AI models in a way that is not harmful to anyone, understanding the responsible angle of AI is important.
In Part 1, we discuss the journey and contributions of nine leaders behind several AI revolutions that TIME missed out on
And AI’s dear children are the reason.
India’s space tech landscape saw remarkable growth, with approximately $205 million raised in funding across 30+ deals between 2014 and July 2023
Jensen Huang of NVIDIA said that Huawei is one of the most technologically advanced companies in the world, and deserves all the praise
Has Silicon Valley really lost its culture of innovation or only Sam Altman feels this way?
“It would take at least two decades for quantum computing to be actually useful, or even exist,” Jensen Huang told AIM
It hurts.
While others have picked up millions of dollars for an idea without a product in hand, Midjourney is doing it the old-fashioned way
“We are going to bring out the fastest computers in the world. These computers are not even in production [so far]. India will be one of the first countries in the world [to get them],” says Jensen Huang, in a recent interaction with AIM.
© Analytics India Magazine Pvt Ltd & AIM Media House LLC 2023
