Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,236 papers · 148 categories

Trend · papers per month

1.2%2.5%3.7%4.9% · May 202619922001200920182026
48 results for forward thinking

Train deep neural networks one layer at a time for faster training and flexibility.

problem Training deep neural networks efficiently and with diverse layers.
method Train layers sequentially, mapping input data forward through each layer.
result Achieved state-of-the-art accuracy on MNIST dataset for convolutional neural networks.

Thinking LLMs struggle with stock prediction, especially as data complexity increases.

problem Evaluating the performance of 'thinking' LLMs in stock prediction, especially under varying levels of cross-sectional complexity.
method Rolling 48m/1m walk-forward evaluation, comparing direct LLMs, TLLMs, and classical learners on cross-sectional ranking loss, MSE, and backtests with transaction costs.
result TLLMs' ranking quality deteriorates as cross-sectional complexity grows, while direct LLMs remain stable.

Breiman's paper sparked debate on the future of statistics and machine learning.

problem The tension between traditional statistical modeling and model-free machine learning approaches.
method Discussion of the implications of machine learning's success and the need for new inferential approaches.
result The importance of understanding 'why' and 'if' questions in machine learning is now recognized.

New method uses reverse thinking to correct machine learning errors.

problem Machine learning methods form inertial thinking schemes that can lead to errors when testing data are vastly different.
method Proposes a new method that uses reverse thinking to correct illusion inertial thinking in machine learning.
result Increases the generalization ability of machine learning methods.

Automatic detection of reflective thinking in math problem solving using body movement data.

problem Detecting reflective thinking in children's mathematical problem solving.
method WeDraw-1 Movement Dataset, Long Short-Term Memory neural networks, end-to-end detection.
result Average F1 scores of 0.73 for automatic detection and 0.79 for end-to-end detection of reflective thinking.

CausalGame benchmarks LLM agents' causal thinking in games.

problem Evaluating causal thinking in AI Scientists with LLMs.
method Interactive games with 14 scenarios incorporating selection bias, measurement error, and hidden confounders.
result None of the 30 LLM agents demonstrated reliable causal thinking, with the best model achieving only 68.0% survival.

This paper uses counterfactual thinking to improve multi-agent reinforcement learning.

problem Improving decision-making in multi-agent environments.
method Proposes a deep reinforcement learning model with counterfactual thinking to generate and evaluate multiple actions.
result Counterfactual thinking enhances agents' performance in multi-agent environments, leading to higher rewards and fair competition.

The paper critiques existing uncertainty concepts and proposes a new decision-theoretic approach.

problem Incoherence in existing discussions of aleatoric and epistemic uncertainty.
method Decision-theoretic perspective that relates uncertainty, predictive performance, and statistical dispersion.
result Popular information-theoretic quantities can be poor estimators but still useful for guiding data acquisition.

A new reinforcement learning method for robots thinking and moving simultaneously.

problem Concurrent control in robotic systems where actions must be decided while the system is still evolving.
method Continuous-time Bellman equations, discretization aware of system delays, and architectural extension to deep reinforcement learning.
result The method successfully handles tasks requiring simultaneous decision-making and action execution.

New method shows how order of gradient updates impacts stability and convergence in deep learning.

problem Training deep learning models can be unstable and computationally expensive.
method Theoretical analysis and experiments with backward-SGD.
result The order of gradient updates affects stability and convergence, leading to improved performance.

The paper explores how innovative financing solutions boost Moroccan businesses' performance.

problem Market volatility, ecological transitions, and technological change pose challenges to business sustainability.
method Examines innovative financing solutions like venture capital, green finance, crowdfunding, and blockchain.
result Embracing innovative financial strategies can transform business challenges into opportunities.

Study shows activist board representation improves Japanese companies' performance.

problem Lack of innovation and improvement in Japanese companies.
method Examined two Japanese companies with activist board representation, analyzing performance metrics.
result Companies with activist board representation experienced significant improvements in stock returns and operational metrics.

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.

DGP learns speech recognition by modeling complex relationships between utterances.

problem Modeling complex relationships in speech recognition without relational data.
method Bayesian nonparametric deep learning method (DGP) that generates infinite probabilistic graphs.
result DGP successfully infers relationships among utterances without relational data during training.

Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats humans …

2016-04-01abs ↗pdf ↗

Simple model outperforms neural networks on language understanding tasks.

problem Neural networks struggle with creating novel expressions from familiar ones.
method Attention-inspired modification of a baseline model, focusing on sequential thinking and acting.
result Simple model achieves good performance on gSCAN tasks, validating the benchmark.

Deep neural networks create surrogate models for high-dimensional uncertainty quantification.

problem Uncertainty quantification for systems with many input parameters is computationally infeasible.
method Constructing a cheap-to-evaluate surrogate model using deep neural networks (DNN) to replace a forward model solver.
result DNN surrogate models can learn a map between an arbitrary snapshot of the diffusion field and the response, overcoming traditional SPDE problem constraints.

TSDS framework reduces edge LLM agent compute by 43%-73% while maintaining safety and reliability.

problem Managing reasoning budget and uncertainty in edge LLM agents.
method Integrates a lightweight convergence probe and a perplexity-based deferral rule calibrated via multi-objective LTT.
result Reduces per-episode thinking compute by 43%-73% over deferral-only baselines.

Research suggests using deep learning for better recommendation systems.

problem Recommender systems rely on proxies for A/B testing, leading to random success.
method Advocates for using deep learning to improve recommendation performance.
result Deep learning can potentially optimize reward in recommendation systems.

In this paper we study n-composition series of affine manifolds. One composition series are classified using gerbe theory. It is natural to think that n-composition series must be classified using n-gerbe theory. In the last section of this, we propose a notion of abelian n-gerbe theory

2001-05-24abs ↗pdf ↗

The paper converts metric bounds to distance function Hölder bounds and proves compactness theorems.

problem Proving geometric stability results with scalar curvature bounds.
method Transforming LpL^p bounds to Hölder bounds for distance functions.
result Compactness theorems and convergence guarantees for Riemannian manifolds.

New models explain residual and dilated dense neural networks using sparse coding.

problem Lack of theoretical understanding of residual and dilated dense neural networks.
method Proposed Res-CSC and MSD-CSC models, derived mathematical relationships, implemented ISTA.
result Mathematical understanding of residual and dilated dense neural networks.

We argue that the present crisis and stalling economy continuing since 2007 are rooted in the delusionary belief in policies based on a "perpetual money machine" type of thinking. We document strong evidence that, since the early 1980s, consumption has been increasingly funded by smaller savings, booming financial prof…

2012-12-12abs ↗pdf ↗

DRNets combine deep learning and reasoning for complex tasks.

problem Solving complex tasks, especially in scientific discovery, with limited supervision.
method DRNets integrate logic and neural network optimization to encode structured latent spaces constrained by prior knowledge.
result DRNets outperform state-of-the-art models in scientific discovery tasks, recovering more precise crystal structures.

We prove new results on existence of solutions for the prescribed gaussian curvature problem on the euclidean sphere S^2. Those results are achieved by relating this problem with the holomorphic triples theory on Riemann surfaces. We think this approach might be applied to study some other semi-linear elliptic equation…

2015-03-19abs ↗pdf ↗

Random neural networks produce functions with a number of knots equal to the number of neurons.

problem Understanding why neural networks with many parameters do not overfit early in training.
method Analyzing random scalar-input feed-forward rectified linear unit architectures, showing they are random linear splines.
result The number of knots in random neural networks is equal to the number of neurons, to very close approximation.

New method uses causal thinking to make AI fairer decisions.

problem Designing fair machine learning models that treat equal individuals equally and unequals unequally.
method Rank-preserving interventional distributions and warping method.
result Warping method effectively identifies discriminated individuals and mitigates unfairness.

We prove some general results about quasi-actions on trees and define Property (QFA), which is analogous to Serre's Property (FA), but in the coarse setting. This property is shown to hold for a class of groups, including SL(n,Z)SL(n,\Z) for n3n\geq 3. We also give a way of thinking about Property (QFA) by breaking it down …

2003-10-06abs ↗pdf ↗

TAMA uses LMMs to detect and interpret anomalies in time series data with few labels.

problem Challenges in manual feature engineering and extensive labeled training data for TSAD.
method Leverages LMMs to convert time series into visual formats for few-shot in-context learning.
result Consistently outperforms state-of-the-art methods in TSAD tasks.