just now

Liquidity Finder Ltd is incorporated in England and Wales, company number 10610740, registered address 167-169 Great Portland Street, Fifth Floor, London W1W 5PF, United Kingdom.
Published: just now

March 30, 2023 - Bloomberg has released a research paper detailing the development of BloombergGPT, a new large-scale generative artificial intelligence (AI) model. This large language model (LLM) has been specifically trained on a wide range of financial data, collected over 40 years by Bloomberg, to support a diverse set of natural language processing (NLP) tasks within the financial industry.
Bloomberg's announcement in full -
Recent advances in Artificial Intelligence (AI) based on LLMs have already demonstrated exciting new applications for many domains. However, the complexity and unique terminology of the financial domain warrant a domain-specific model. BloombergGPT represents the first step in the development and application of this new technology for the financial industry. This model will assist Bloomberg in improving existing financial NLP tasks, such as sentiment analysis, named entity recognition, news classification, and question answering, among others. Furthermore, BloombergGPT will unlock new opportunities for marshalling the vast quantities of data available on the Bloomberg Terminal to better help the firm’s customers, while bringing the full potential of AI to the financial domain.
For more than a decade, Bloomberg has been a trailblazer in its application of AI, Machine Learning, and NLP in finance. Today, Bloomberg supports a very large and diverse set of NLP tasks that will benefit from a new finance-aware language model. Bloomberg researchers pioneered a mixed approach that combines both finance data with general-purpose datasets to train a model that achieves best-in-class results on financial benchmarks, while also maintaining competitive performance on general-purpose LLM benchmarks.
To achieve this milestone, Bloomberg’s ML Product and Research group collaborated with the firm’s AI Engineering team to construct one of the largest domain-specific datasets yet, drawing on the company’s existing data creation, collection, and curation resources. As a financial data company, Bloomberg’s data analysts have collected and maintained financial language documents over the span of forty years. The team pulled from this extensive archive of financial data to create a comprehensive 363 billion token dataset consisting of English financial documents.
This data was augmented with a 345 billion token public dataset to create a large training corpus with over 700 billion tokens. Using a portion of this training corpus, the team trained a 50-billion parameter decoder-only causal language model. The resulting model was validated on existing finance-specific NLP benchmarks, a suite of Bloomberg internal benchmarks, and broad categories of general-purpose NLP tasks from popular benchmarks (e.g., BIG-bench Hard, Knowledge Assessments, Reading Comprehension, and Linguistic Tasks). Notably, the BloombergGPT model outperforms existing open models of a similar size on financial tasks by large margins, while still performing on par or better on general NLP benchmarks.
“For all the reasons generative LLMs are attractive – few-shot learning, text generation, conversational systems, etc. – we see tremendous value in having developed the first LLM focused on the financial domain,” said Shawn Edwards, Bloomberg’s Chief Technology Officer. “BloombergGPT will enable us to tackle many new types of applications, while it delivers much higher performance out-of-the-box than custom models for each application, at a faster time-to-market.”
“The quality of machine learning and NLP models comes down to the data you put into them,” explained Gideon Mann, Head of Bloomberg’s ML Product and Research team. “Thanks to the collection of financial documents Bloomberg has curated over four decades, we were able to carefully create a large and clean, domain-specific dataset to train a LLM that is best suited for financial use cases. We’re excited to use BloombergGPT to improve existing NLP workflows, while also imagining new ways to put this model to work to delight our customers.”
To learn more about the development of BloombergGPT, read the paper on arXiv here.
There is a good overview here of BloombergGPT in a 4-minute video, from Richard walker at Lucidate:
We're the largest marketplace to connect with brokers, Fintech companies & digital asset firms. Want to partner? Let's get in touch.
Select the categories and companies you wish to follow directly to your person rss feed.
Create Custom RSS FeedSign up and join over 5,000 professional members who receive personalized news alerts, curated professional connections, and more for free!
Why did gold keep falling after reaching $3,880? See how a failed 4-hour retest confirmed weakness in XAUUSD and why $3,400 is the next weekly support to watch.
Reality television personality Aaron Chalmers and finfluencer Jason Agbodza have been fined after pleading guilty to illegally promoting an unauthorised foreign exchange trading scheme on social media.
UK retail sales rose a stronger-than-expected 0.7% in October, giving sterling some support, while a deepening tech sell-off pushed the dollar lower overnight against all major currencies. Thanim Islam, Head of FX Analysis at Equals, says the dollar remains in charge as the week ends, with the Fed expected to raise rates more than both the Bank of England and the ECB.
Nvidia beats expectations, reporting $115.8bn in revenue and raising its long-term growth outlook, but the stock still struggled to sustain its gains. AI remains fundamentally strong, but investor expectations are becoming more demanding, particularly around capex and profitability. Wall Street suffered a major sell-off, with the S&P 500 entering a six-session losing streak as rate concerns intensified.
iSAM Securities reviews the ris of weekend gold trading
Webull Corporation reported record third quarter 2026 revenue of $218.7 million, up 39% year-on-year, with customer assets reaching $32.4 billion and options contracts volume up 79%. Net income attributable to the company was $31.2 million, while margin debit balances passed $1 billion for the first time.
Anthropic, the company behind the Claude family of AI models, begins trading on the Nasdaq today under the ticker ANT. It priced its shares at $83 last night, above its marketed range of $70 to $78, and raised about $81.3 billion. That makes it the largest initial public offering on record, ahead of the roughly $75 billion SpaceX raised in June. This note covers what investors need to know: the terms, the business, how the company is governed, and what to watch.
Nvidia's first results without Jensen Huang answered the question that has hung over the AI trade since October. Revenue of $115.8 billion beat consensus by about $7 billion, fourth-quarter guidance of $130 billion came in roughly $5 billion above expectations, and the company raised its outlook for next fiscal year to around 75% growth. All of this assumes no data-centre sales to China.
Marex has entered the football financing market through a partnership with specialist sports investment and advisory firm Pinto Capital, targeting clubs in the Premier League and EFL Championship.
Around 110,000 UK retail investors have secured $300 million worth of Anthropic shares in its record-breaking $81 billion IPO. The heavily oversubscribed offer, coordinated by RetailBook, saw smaller investors prioritised, with 63% of applicants receiving their full allocation.