Compact models match or exceed GPT's performance in financial news sentiment analysis.
problem Improving financial sentiment analysis models without large computational costs.
method Fine-tuning non-generative, small-sized models (FinBERT, FinDRoBERTa) on a novel market score database.
result Fine-tuned models outperform GPT-3.5 and GPT-4 in zero-shot learning for financial news sentiment analysis.
Large financial dataset tracks FOMC communications and their impact.
problem Understanding how FOMC communications influence financial markets.
method Constructed a large annotated dataset of FOMC speeches, minutes, and transcripts. Developed a hawk-dove classification task. Evaluated various models on the dataset and used RoBERTa-large for monetary policy stance measurement.
result Monetary policy stance measures derived from FOMC documents predict market performance.
NIFTY dataset for financial forecasting models.
problem Improving financial market forecasting with LLMs.
method Two datasets for supervised and reinforcement learning.
result Demonstrates applications in stock price prediction.
Generative models create synthetic MRI brain scans for research.
problem Lack of large, diverse medical datasets for machine learning.
method Training generative models on 40,000 MRI scans to produce 18,000 synthetic samples.
result Generated samples improve machine learning model accuracy.
Classifier detects LLM-generated text with guarantees.
problem Detecting fake LLM-generated text to prevent misuse.
method Trained classifier without auxiliary info, distinguishing human and LLM text.
result Achieves higher accuracy than existing detectors with type-I error control.
Method constructs finance LLMs without instruction data using pretraining and model merging.
problem Developing domain-specific LLMs for finance is resource-intensive.
method Continual pretraining on financial data + model merging of instruction-tuned and domain-specific pretrained vectors.
result Successfully constructs instruction-tuned LLMs for finance without additional instruction data.
FairJob dataset for job recommendations in advertising, preserving fairness and utility.
problem Fairness in online job recommendations for advertising.
method Anonymized dataset with proxy for sensitive attributes, fairness metric computation.
result Demonstrated potential improvements in fairness with trade-offs in utility.
Model learns brevity by exposing to easy problems, improving efficiency without explicit length penalties.
problem Excessive verbosity in step-by-step reasoning models trained with RLVR.
method Retaining and up-weighting moderately easy problems as implicit length regularizers.
result Model generates solutions that are, on average, nearly twice as short without explicit length penalties.
Unified framework deciphers global central bank communications.
problem Misinterpretations of central bank communications can disproportionately impact vulnerable populations.
method Developed the World Central Banks (WCB) dataset, annotated and reviewed sentences, defined tasks, and benchmarked models.
result A model trained on aggregated data across banks outperforms models trained on individual bank data.
LLM Pro Finance Suite enhances financial NLP with instruction-tuned models.
problem Limited NLP capabilities for financial tasks in generalist models.
method Instruction-tuned large language models fine-tuned on financial data.
result Consistent improvement over state-of-the-art baselines in finance tasks.
A new Hopfield model reduces inefficiency in large transformer models.
problem Inefficiency in training large transformer-based models.
method Introduces an Outlier-Efficient Hopfield Model (OutEffHop) to improve model performance.
result Achieves up to 26% reduction in model output norms across four models.
ISOMORPH creates a digital twin for supply chain logistics, advancing time-series forecasting benchmarks.
problem Lack of public benchmarks for supply chain logistics time-series forecasting.
method Developed a digital twin simulator with interpretable parameters and modular topology, generating datasets and verifying conservation laws.
result Foundation models achieve MASE values exceeding public benchmarks at low-to-moderate horizons, supporting UQ.
G1 uses RL to enhance LLMs' graph reasoning, improving performance on diverse tasks.
problem Limited graph reasoning abilities of LLMs, especially in synthetic graph-theoretic tasks.
method Curated synthetic graph dataset, RL training on LLMs.
result Significant improvements in graph reasoning, zero-shot generalization to unseen tasks.
SPIN converts weak LLMs to strong ones using self-play.
problem Growing strong LLMs without human-annotated data.
method Self-play fine-tuning (SPIN) starting from a supervised fine-tuned model.
result SPIN significantly improves LLM performance across benchmarks.