FinBERT
PulseAugur coverage of FinBERT — every cluster mentioning FinBERT across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
Financial NLP benchmarks suffer from temporal leakage, inflating performance metrics
A new audit of financial news NLP benchmarks reveals significant temporal leakage, where random train-test splits inflate performance metrics by up to 6.5x compared to chronological splits. This leakage is particularly …
-
New RA-FinBERT model boosts financial sentiment analysis with rule-based features
Researchers have developed RA-FinBERT, a novel framework for financial sentiment analysis that enhances accuracy by integrating rule-derived features with a pre-trained language model. This approach combines sentiment p…
-
LLMs Show High Accuracy in Financial Sentiment, But Fail to Predict Stock Returns
A new study benchmarks several large language models (LLMs) for their effectiveness in financial sentiment classification and return predictability. Researchers found that while models like Mistral-7B and QLoRA-adapted …
-
LLaMA-3.1-70B extracts structured data from financial news to boost stock prediction
Researchers have developed a new framework for extracting structured information from financial news, moving beyond traditional sentiment analysis. This framework utilizes LLaMA-3.1-70B to identify six semantic dimensio…
-
AI system AWARE-FX quantifies FX hedging disclosures in corporate reports
Researchers have developed AWARE-FX, an AI system designed to analyze corporate annual reports and quantify foreign-exchange hedging disclosures. This system integrates a specialized lexicon, logic for negation and acco…
-
New AI Model RAML Improves Bitcoin Price Prediction Using Dynamic Sentiment Fusion
Researchers have developed a new model called Regime-Aware Multi-Modal Learning (RAML) to predict Bitcoin price movements on sub-daily timescales. Unlike traditional methods that statically combine price and social sent…
-
TriAgent cuts LLM costs for financial sentiment analysis
Researchers have developed TriAgent, a novel multi-agent system designed to reduce the cost of financial sentiment analysis using large language models. The system stratifies agents by contextual granularity, employing …
-
Domain adaptation efficacy depends on pre-trained model's domain knowledge
A new study investigates the effectiveness of domain adaptation techniques when using frozen pre-trained language model backbones for sentiment analysis. The research evaluated different adaptation methods like DANN, MM…
-
New AI Framework Enhances Audit Risk Assessment with Uncertainty Modeling
Researchers have developed UMAR, a novel multi-agent framework designed to improve audit risk assessment by explicitly modeling uncertainty and evidence conflict. UMAR utilizes three specialized agents—MD&A Text Agent, …
-
New method uses FinBERT embeddings for better stock market prediction
Researchers have developed a new method to improve financial forecasting by using high-dimensional embeddings from FinBERT instead of simple sentiment scores. Their Transformer-based architecture, which incorporates Sia…
-
Deep Learning Models Predict Stock Price Direction Using Multi-modal Data
Researchers have developed a multi-modal deep learning approach to predict stock price direction on earnings announcement days. The study combines fundamental metrics, technical indicators, and sentiment analysis from f…
-
Researchers aggregate zero-shot LLM outputs for better stock return prediction
Researchers have developed a lightweight supervised aggregator to combine outputs from multiple zero-shot Large Language Models (LLMs) for classifying corporate disclosures. This method aims to improve prediction accura…