Institution profile

Warsaw University of Technology

Academic institutioneurope · pl
Official website
Research library302linked papers
Opportunities0open roles
Selected work

Representative Papers

A Finite Difference Approximation of Second Order Regularization of Neural-SDFs

Nov 12, 2025

To address the high computational and memory overhead of curvature regularization in neural signed distance field (SDF) learning—stemming from reliance on second-order automatic differentiation—this paper proposes a lightweight finite-difference-based regularization framework. We introduce, for the first time, an O(h²)-accurate finite-difference stencil for explicit SDF curvature modeling, bypassing Hessian construction and second-order gradients entirely. The method enables plug-and-play approximations of both Gaussian curvature and rank-deficiency loss. Empirically, it matches the reconstruction accuracy of automatic-differentiation-based curvature regularization while reducing GPU memory consumption and training time by up to 50%. Moreover, it demonstrates strong robustness to sparse, incomplete, and non-CAD data. Our core contribution is achieving high-fidelity geometric regularization at the cost of only low-order differentiation, thereby significantly improving the efficiency and scalability of neural SDF learning.

1 citationsRead paper

Integrating Traditional Technical Analysis with AI: A Multi-Agent LLM-Based Approach to Stock Market Forecasting

Jun 20, 2025International Conference on Agents and Artificial Intelligence

Traditional technical analysis suffers from limited predictive accuracy in financial markets characterized by nonlinearity, high noise, and frequent exogenous shocks. To address this, we propose ElliottAgents—a novel interpretable AI trading system that is the first to deeply integrate Elliott Wave Theory into a large language model (LLM)-based multi-agent framework, synergizing retrieval-augmented generation (RAG), deep reinforcement learning (DRL), and multi-agent coordination. Our system enables automated wave-pattern recognition, cross-temporal trend inference, and decision traceability, thereby bridging interpretability and adaptability in AI-driven trading. Empirical evaluation on historical U.S. equity data demonstrates that ElliottAgents significantly outperforms baseline methods: waveform identification accuracy and trend prediction stability are markedly improved, with signal accuracy increasing by 23.6% and strategy Sharpe ratio rising by 31.4%.

1 citationsRead paper

Applying Informer for Option Pricing: A Transformer-Based Approach

Jun 05, 2025International Conference on Agents and Artificial Intelligence

Option pricing faces challenges in modeling market nonlinearity, time-varying volatility, and long-range dependencies. Conventional approaches—such as the Black–Scholes model and LSTM-based methods—exhibit limited robustness and adaptability to rapidly evolving financial dynamics. This paper introduces, for the first time, the lightweight and efficient Informer architecture to option pricing. Leveraging ProbSparse self-attention, distilling encoders, and a generative decoder, Informer effectively captures the dynamic structure of high-frequency, heterogeneous financial time series. The proposed data-driven framework significantly enhances real-time responsiveness to market regime shifts and improves generalization capability. Empirical evaluations across multiple markets demonstrate that the method reduces average pricing error by 37% relative to both Black–Scholes and LSTM baselines. Moreover, it achieves superior prediction stability and markedly improved cross-maturity and cross-contract generalization. This work establishes a novel paradigm for high-accuracy, low-latency derivative pricing.

1 citationsRead paper
Recent publications

Latest Papers

Solving versus Verifying: Catching Contradictions in Tax Reasoning Systems

Sep 05, 2026

Large language models now compute correct tax liabilities on over 90% of well-formed cases in statutory benchmarks, which makes them candidates for the tax-advisory and compliance systems that consume such an answer directly. Real legal inputs, however, are frequently defective: required facts are missing, or stated facts contradict one another. Accuracy on clean benchmarks says nothing about how a model behaves then, and a system that computes straight through a defective input returns a confident number with no sign that anything is wrong. This raises two questions: does a model asked to solve a case abstain when the input is defective, and when it does not, can the same model catch the defect when asked instead to verify the input? We study six recent models on SARA-derived tax cases under missing-fact and contradictory-fact perturbations. The strongest models abstain when a fact is missing but compute through injected contradictions, returning the clean-input answer 63-76% of the time with no signal of the conflict; asked instead to verify the same input, they flag most of those contradictions. We wire that verification call into a simple contradiction gate: one extra call that abstains when the model reports a conflict. Across all six models it recovers most of the missed contradiction abstention at a clean-accuracy cost of at most about 5 percentage points, with no training and no external tooling. High accuracy on well-formed inputs is therefore an incomplete measure of reliability, and the detection the solver misses is cheaply recoverable with a single self-check.

0 citationsRead paper