Finance

TL;DR A new paper shows how to insert a sparse, interpretable layer into an LLM to expose plain‑English concepts (e.g., sentiment, risk, timing) and steer them like dials without retraining. In finance news prediction, these interpretable features outperform final‑layer embeddings and reveal that sentiment, market/technical cues, and timing drive most short‑horizon alpha. Steering also debiases optimism, lifting Sharpe by nudging the model negative on sentiment. Why this matters (and what’s new) Finance teams have loved LLMs’ throughput but hated their opacity. This paper demonstrates a lightweight path to transparent performance: ...

TL;DR FinCast is a 1B‑parameter, decoder‑only Transformer trained on >20B financial time points with a token‑level sparse Mixture‑of‑Experts (MoE), learnable frequency embeddings, and a Point‑Quantile (PQ) loss that combines Huber point forecasts with quantile targets and a trend‑consistency term. In zero‑shot benchmarks across crypto/FX/stocks/futures, it reports ~20% lower MSE vs leading generic time‑series FMs, and it also beats supervised SOTAs—even without fine‑tuning—then widens the gap with a light fine‑tune. If you build risk or execution systems, the interesting part isn’t just accuracy points; it’s the shape of the predictions (tail‑aware, regime‑sensitive) and the deployment economics (conditional compute via sparse MoE + patching). ...

If AutoML is a fast car, financial institutions need a train with tracks—a workflow that knows where it’s going, logs every switch, and won’t derail when markets regime-shift. A new framework called TS-Agent proposes exactly that: a structured, auditable, LLM-driven agent that plans model development for financial time series instead of blindly searching. Unlike generic AutoML, TS-Agent formalizes modeling as a multi-stage decision process—Model Pre-selection → Code Refinement → Fine-tuning—and anchors each step in domain-curated knowledge banks and reflective feedback from real runs. The result is not just higher accuracy; it’s traceability and consistency that pass governance sniff tests. ...

Financial AI promises speed and scale — but in finance, a single misplaced digit can be the difference between compliance and catastrophe. The FAITH (Framework for Assessing Intrinsic Tabular Hallucinations) benchmark tackles this risk head‑on, probing how well large language models can faithfully extract and compute numbers from the dense, interconnected tables in 10‑K filings. From Idea to Dataset: Masking With a Purpose FAITH reframes hallucination detection as a context‑aware masked span prediction task. It takes real S&P 500 annual reports, hides specific numeric spans, and asks the model to recover them — but only after ensuring three non‑negotiable conditions: ...

In asset management, few debates are more charged than the tug-of-war between causal purity and predictive utility. For years, a growing number of voices in empirical finance have argued that causal factor models are a necessary condition for portfolio efficiency. If a model omits a confounder, the logic goes, directional failure and Sharpe ratio collapse are inevitable. But what if this is more myth than mathematical law? A recent paper titled “The Myth of Causal Necessity” by Alejandro Rodriguez Dominguez delivers a sharp counterpunch to this orthodoxy. Through formal derivations and simulation-based counterexamples, it exposes the fragility of the causal necessity argument and makes the case that predictive models can remain both viable and efficient even when structurally misspecified. ...

Lévy processes — stochastic processes with jumps — are the bedrock of modern financial modeling. From the Variance Gamma model to the CGMY framework, these models have replaced Brownian motion in capturing the reality of financial returns: asymmetry, fat tails, and sudden discontinuities. But what if we told you these processes don’t just live on probability distributions — they live on manifolds? ...

Finance may seem like the crown jewel of modern institutions—replete with contracts, algorithms, and global markets. But what if its deepest logic predates banks, money, and even language? In a compelling new paper, Finance as Extended Biology (arXiv:2506.00099), Egil Diau argues that the cognitive substrate of finance is not institutional architecture but reciprocity—a fundamental behavioral mechanism observed in primates and ancient human societies alike. Credit, insurance, token exchange, and investment, he contends, are not designed structures but emergent transformations of this ancient cooperative logic. ...

Deep learning has revolutionized many domains of finance, but when it comes to asset pricing, its power is often undercut by a familiar enemy: noise. Financial datasets are notoriously riddled with weak signals and irrelevant patterns, which easily mislead even the most sophisticated models. The result? Overfitting, poor generalization, and ultimately, bad bets. A recent paper by Che Sun proposes an elegant fix by drawing inspiration from information theory. Titled An Information Bottleneck Asset Pricing Model, the paper integrates information bottleneck (IB) regularization into an autoencoder-based asset pricing framework. The goal is simple yet profound: compress away the noise, and preserve only what matters for predicting asset returns. ...

When ChatGPT makes up a statistic or misstates a date, it’s annoying. But when a financial assistant claims the wrong interest expense or misattributes a revenue source, it could move markets or mislead clients. This is the stark reality FRED confronts head-on. FRED—short for Financial Retrieval-Enhanced Detection and Editing—is a framework fine-tuned to spot and fix factual errors in financial LLM outputs. Developed by researchers at Pegasi AI, it isn’t just another hallucination detection scheme. It’s an auditor with a domain-specific brain. ...

In quantitative finance, sparse portfolio optimization is a famously unforgiving problem. Selecting the top m assets from a universe of n under budget and risk constraints is NP-hard, highly sensitive to hyperparameters, and often brittle in volatile markets. Traditional solutions—from greedy algorithms to convex relaxations—either crumble under market shifts or produce opaque, overfitted outputs. But what if we reframed the problem entirely? Enter EFS (Evolutionary Factor Search), a radical new framework that turns sparse portfolio construction into an LLM-guided ranking game. Instead of laboriously tuning machine learning models or relying on rigid heuristics, EFS lets large language models generate, evolve, and select alpha factors—and it does so in a way that is not just automated, but interpretable, adaptive, and surprisingly effective. ...