FinBERT adapts BERT to the financial domain and outperforms prior state-of-the-art methods on financial sentiment analysis tasks.
Denis Gromb and Dimitri Vayanos
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
verdicts
UNVERDICTED 5representative citing papers
Prediction market prices for Bitcoin thresholds on Polymarket exceed option-implied probabilities from Binance by 5.6-6.3 percentage points on average, with the gap persistent yet mean-reverting.
A new model derives a convex systemic risk coupling r(φ) that grows superlinearly with AI adoption share, producing a saddle-node bifurcation to algorithmic monoculture and 18-54% tail-loss amplification, validated on SEC 13F holdings data.
Value of information to informed traders equals price-order flow covariance and totals 0.04% of market cap, much less than active management fees.
The paper develops a four-layer AI agent architecture and the Agentic Financial Market Model linking agent parameters such as autonomy and coupling to market efficiency, liquidity, and systemic risk, with an exploratory event-study application.
citing papers explorer
-
Do Prediction Markets Match Option Prices? Bitcoin Threshold Evidence from Binance and Polymarket
Prediction market prices for Bitcoin thresholds on Polymarket exceed option-implied probabilities from Binance by 5.6-6.3 percentage points on average, with the gap persistent yet mean-reverting.
-
Artificial Intelligence and Systemic Risk: A Unified Model of Performative Prediction, Algorithmic Herding, and Cognitive Dependency in Financial Markets
A new model derives a convex systemic risk coupling r(φ) that grows superlinearly with AI adoption share, producing a saddle-node bifurcation to algorithmic monoculture and 18-54% tail-loss amplification, validated on SEC 13F holdings data.
-
The Value of Information: A Puzzle
Value of information to informed traders equals price-order flow covariance and totals 0.04% of market cap, much less than active management fees.
-
AI Agents in Financial Markets: Architecture, Applications, and Systemic Implications
The paper develops a four-layer AI agent architecture and the Agentic Financial Market Model linking agent parameters such as autonomy and coupling to market efficiency, liquidity, and systemic risk, with an exploratory event-study application.