Multimodal Language Models with Modality-Specific Experts for Financial Forecasting from Interleaved Sequences of Text and Time Series ArXiv ID: 2509.19628 “View on arXiv” Authors: Ross Koval, Nicholas Andrews, Xifeng Yan Abstract Text and time series data offer complementary views of financial markets: news articles provide narrative context about company events, while stock prices reflect how markets react to those events. However, despite their complementary nature, effectively integrating these interleaved modalities for improved forecasting remains challenging. In this work, we propose a unified neural architecture that models these interleaved sequences using modality-specific experts, allowing the model to learn unique time series patterns, while still enabling joint reasoning across modalities and preserving pretrained language understanding capabilities. To further improve multimodal understanding, we introduce a cross-modal alignment framework with a salient token weighting mechanism that learns to align representations across modalities with a focus on the most informative tokens. We demonstrate the effectiveness of our approach on a large-scale financial forecasting task, achieving state-of-the-art performance across a wide variety of strong unimodal and multimodal baselines. We develop an interpretability method that reveals insights into the value of time series-context and reinforces the design of our cross-modal alignment objective. Finally, we demonstrate that these improvements translate to meaningful economic gains in investment simulations. ...

Sharp Large Deviations and Gibbs Conditioning for Threshold Models in Portfolio Credit Risk ArXiv ID: 2509.19151 “View on arXiv” Authors: Fengnan Deng, Anand N. Vidyashankar, Jeffrey F. Collamore Abstract We obtain sharp large deviation estimates for exceedance probabilities in dependent triangular array threshold models with a diverging number of latent factors. The prefactors quantify how latent-factor dependence and tail geometry enter at leading order, yielding three regimes: Gaussian or exponential-power tails produce polylogarithmic refinements of the Bahadur-Rao $n^{"-1/2"}$ law; regularly varying tails yield index-driven polynomial scaling; and bounded-support (endpoint) cases lead to an $n^{"-3/2"}$ prefactor. We derive these results through Laplace-Olver asymptotics for exponential integrals and conditional Bahadur-Rao estimates for the triangular arrays. Using these estimates, we establish a Gibbs conditioning principle in total variation: conditioned on a large exceedance event, the default indicators become asymptotically i.i.d., and the loss-given-default distribution is exponentially tilted (with the boundary case handled by an endpoint analysis). As illustrations, we obtain second-order approximations for Value-at-Risk and Expected Shortfall, clarifying when portfolios operate in the genuine large-deviation regime. The results provide a transferable set of techniques-localization, curvature, and tilt identification-for sharp rare-event analysis in dependent threshold systems. ...

Enhanced fill probability estimates in institutional algorithmic bond trading using statistical learning algorithms with quantum computers ArXiv ID: 2509.17715 “View on arXiv” Authors: Axel Ciceri, Austin Cottrell, Joshua Freeland, Daniel Fry, Hirotoshi Hirai, Philip Intallura, Hwajung Kang, Chee-Kong Lee, Abhijit Mitra, Kentaro Ohno, Das Pemmaraju, Manuel Proissl, Brian Quanz, Del Rajan, Noriaki Shimada, Kavitha Yograj Abstract The estimation of fill probabilities for trade orders represents a key ingredient in the optimization of algorithmic trading strategies. It is bound by the complex dynamics of financial markets with inherent uncertainties, and the limitations of models aiming to learn from multivariate financial time series that often exhibit stochastic properties with hidden temporal patterns. In this paper, we focus on algorithmic responses to trade inquiries in the corporate bond market and investigate fill probability estimation errors of common machine learning models when given real production-scale intraday trade event data, transformed by a quantum algorithm running on IBM Heron processors, as well as on noiseless quantum simulators for comparison. We introduce a framework to embed these quantum-generated data transforms as a decoupled offline component that can be selectively queried by models in low-latency institutional trade optimization settings. A trade execution backtesting method is employed to evaluate the fill prediction performance of these models in relation to their input data. We observe a relative gain of up to ~ 34% in out-of-sample test scores for those models with access to quantum hardware-transformed data over those using the original trading data or transforms by noiseless quantum simulation. These empirical results suggest that the inherent noise in current quantum hardware contributes to this effect and motivates further studies. Our work demonstrates the emerging potential of quantum computing as a complementary explorative tool in quantitative finance and encourages applied industry research towards practical applications in trading. ...

FinFlowRL: An Imitation-Reinforcement Learning Framework for Adaptive Stochastic Control in Finance ArXiv ID: 2509.17964 “View on arXiv” Authors: Yang Li, Zhi Chen, Steve Y. Yang, Ruixun Zhang Abstract Traditional stochastic control methods in finance rely on simplifying assumptions that often fail in real world markets. While these methods work well in specific, well defined scenarios, they underperform when market conditions change. We introduce FinFlowRL, a novel framework for financial stochastic control that combines imitation learning with reinforcement learning. The framework first pretrains an adaptive meta policy by learning from multiple expert strategies, then finetunes it through reinforcement learning in the noise space to optimize the generation process. By employing action chunking, that is generating sequences of actions rather than single decisions, it addresses the non Markovian nature of financial markets. FinFlowRL consistently outperforms individually optimized experts across diverse market conditions. ...

Analysis of the Impact of an Execution Algorithm with an Order Book Imbalance Strategy on a Financial Market Using an Agent-based Simulation ArXiv ID: 2509.16912 “View on arXiv” Authors: Shuto Endo, Takanobu Mizuta, Isao Yagi Abstract Order book imbalance (OBI) - buy orders minus sell orders near the best quote - measures supply-demand imbalance that can move prices. OBI is positively correlated with returns, and some investors try to use it to improve performance. Large orders placed at once can reveal intent, invite front-running, raise volatility, and cause losses. Execution algorithms therefore split parent orders into smaller lots to limit price distortion. In principle, using OBI inside such algorithms could improve execution, but prior evidence is scarce because isolating OBI’s effect in real markets is nearly impossible amid many external factors. Multi-agent simulation offers a way to study this. In an artificial market, individual actors are agents whose rules and interactions form the model. This study builds an execution algorithm that accounts for OBI, tests it across several market patterns in artificial markets, and analyzes mechanisms, comparing it with a conventional (OBI-agnostic) algorithm. Results: (i) In stable markets, the OBI strategy’s performance depends on the number of order slices; outcomes vary with how the parent order is partitioned. (ii) In markets with unstable prices, the OBI-based algorithm outperforms the conventional approach. (iii) Under spoofing manipulation, the OBI strategy is not significantly worse than the conventional algorithm, indicating limited vulnerability to spoofing. Overall, OBI provides a useful signal for execution. Incorporating OBI can add value - especially in volatile conditions - while remaining reasonably robust to spoofing; in calm markets, benefits are sensitive to slicing design. ...

Improving S&P 500 Volatility Forecasting through Regime-Switching Methods ArXiv ID: 2510.03236 “View on arXiv” Authors: Ava C. Blake, Nivika A. Gandhi, Anurag R. Jakkula Abstract Accurate prediction of financial market volatility is critical for risk management, derivatives pricing, and investment strategy. In this study, we propose a multitude of regime-switching methods to improve the prediction of S&P 500 volatility by capturing structural changes in the market across time. We use eleven years of SPX data, from May 1st, 2014 to May 27th, 2025, to compute daily realized volatility (RV) from 5-minute intraday log returns, adjusted for irregular trading days. To enhance forecast accuracy, we engineered features to capture both historical dynamics and forward-looking market sentiment across regimes. The regime-switching methods include a soft Markov switching algorithm to estimate soft-regime probabilities, a distributional spectral clustering method that uses XGBoost to assign clusters at prediction time, and a coefficient-based soft regime algorithm that extracts HAR coefficients from time segments segmented through the Mood test and clusters through Bayesian GMM for soft regime weights, using XGBoost to predict regime probabilities. Models were evaluated across three time periods–before, during, and after the COVID-19 pandemic. The coefficient-based clustering algorithm outperformed all other models, including the baseline autoregressive model, during all time periods. Additionally, each model was evaluated on its recursive forecasting performance for 5- and 10-day horizons during each time period. The findings of this study demonstrate the value of regime-aware modeling frameworks and soft clustering approaches in improving volatility forecasting, especially during periods of heightened uncertainty and structural change. ...

LEMs: A Primer On Large Execution Models ArXiv ID: 2509.25211 “View on arXiv” Authors: Remi Genet, Hugo Inzirillo Abstract This paper introduces Large Execution Models (LEMs), a novel deep learning framework that extends transformer-based architectures to address complex execution problems with flexible time boundaries and multiple execution constraints. Building upon recent advances in neural VWAP execution strategies, LEMs generalize the approach from fixed-duration orders to scenarios where execution duration is bounded between minimum and maximum time horizons, similar to share buyback contract structures. The proposed architecture decouples market information processing from execution allocation decisions: a common feature extraction pipeline using Temporal Kolmogorov-Arnold Networks (TKANs), Variable Selection Networks (VSNs), and multi-head attention mechanisms processes market data to create informational context, while independent allocation networks handle the specific execution logic for different scenarios (fixed quantity vs. fixed notional, buy vs. sell orders). This architectural separation enables a unified model to handle diverse execution objectives while leveraging shared market understanding across scenarios. Through comprehensive empirical evaluation on intraday cryptocurrency markets and multi-day equity trading using DOW Jones constituents, we demonstrate that LEMs achieve superior execution performance compared to traditional benchmarks by dynamically optimizing execution paths within flexible time constraints. The unified model architecture enables deployment across different execution scenarios (buy/sell orders, varying duration boundaries, volume/notional targets) through a single framework, providing significant operational advantages over asset-specific approaches. ...

Quantum Adaptive Self-Attention for Financial Rebalancing: An Empirical Study on Automated Market Makers in Decentralized Finance ArXiv ID: 2509.16955 “View on arXiv” Authors: Chi-Sheng Chen, Aidan Hung-Wen Tsai Abstract We formulate automated market maker (AMM) \emph{“rebalancing”} as a binary detection problem and study a hybrid quantum–classical self-attention block, \textbf{“Quantum Adaptive Self-Attention (QASA)”}. QASA constructs quantum queries/keys/values via variational quantum circuits (VQCs) and applies standard softmax attention over Pauli-$Z$ expectation vectors, yielding a drop-in attention module for financial time-series decision making. Using daily data for \textbf{“BTCUSDC”} over \textbf{“Jan-2024–Jan-2025”} with a 70/15/15 time-series split, we compare QASA against classical ensembles, a transformer, and pure quantum baselines under Return, Sharpe, and Max Drawdown. The \textbf{“QASA-Sequence”} variant attains the \emph{“best single-model risk-adjusted performance”} (\textbf{“13.99%”} return; \textbf{“Sharpe 1.76”}), while hybrid models average \textbf{“11.2%”} return (vs.\ 9.8% classical; 4.4% pure quantum), indicating a favorable performance–stability–cost trade-off. ...

Increase Alpha: Performance and Risk of an AI-Driven Trading Framework ArXiv ID: 2509.16707 “View on arXiv” Authors: Sid Ghatak, Arman Khaledian, Navid Parvini, Nariman Khaledian Abstract There are inefficiencies in financial markets, with unexploited patterns in price, volume, and cross-sectional relationships. While many approaches use large-scale transformers, we take a domain-focused path: feed-forward and recurrent networks with curated features to capture subtle regularities in noisy financial data. This smaller-footprint design is computationally lean and reliable under low signal-to-noise, crucial for daily production at scale. At Increase Alpha, we built a deep-learning framework that maps over 800 U.S. equities into daily directional signals with minimal computational overhead. The purpose of this paper is twofold. First, we outline the general overview of the predictive model without disclosing its core underlying concepts. Second, we evaluate its real-time performance through transparent, industry standard metrics. Forecast accuracy is benchmarked against both naive baselines and macro indicators. The performance outcomes are summarized via cumulative returns, annualized Sharpe ratio, and maximum drawdown. The best portfolio combination using our signals provides a low-risk, continuous stream of returns with a Sharpe ratio of more than 2.5, maximum drawdown of around 3%, and a near-zero correlation with the S&P 500 market benchmark. We also compare the model’s performance through different market regimes, such as the recent volatile movements of the US equity market in the beginning of 2025. Our analysis showcases the robustness of the model and significantly stable performance during these volatile periods. Collectively, these findings show that market inefficiencies can be systematically harvested with modest computational overhead if the right variables are considered. This report will emphasize the potential of traditional deep learning frameworks for generating an AI-driven edge in the financial market. ...

Enhancing OHLC Data with Timing Features: A Machine Learning Evaluation ArXiv ID: 2509.16137 “View on arXiv” Authors: Ruslan Tepelyan Abstract OHLC bar data is a widely used format for representing financial asset prices over time due to its balance of simplicity and informativeness. Bloomberg has recently introduced a new bar data product that includes additional timing information-specifically, the timestamps of the open, high, low, and close prices within each bar. In this paper, we investigate the impact of incorporating this timing data into machine learning models for predicting volume-weighted average price (VWAP). Our experiments show that including these features consistently improves predictive performance across multiple ML architectures. We observe gains across several key metrics, including log-likelihood, mean squared error (MSE), $R^2$, conditional variance estimation, and directional accuracy. ...