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.

169,291 papers · 148 categories

Trend · papers per month

count % of corpus

48 results for “Length bias”

Adaptively Truncating Backpropagation Through Time to Control Gradient Biascs.LG

Adaptive TBPTT controls gradient bias in RNNs for faster convergence.

problem Choosing optimal truncation length in TBPTT for RNNs is difficult.

method Adaptive TBPTT converts lag selection to bias control, estimating optimal truncation length during training.

result Adaptive TBPTT improves convergence rate and computational efficiency in RNNs.

TBPTT Recurrent Neural Networks Gradient Bias Adaptivity Convergence Rate

2019-05-17abs ↗pdf ↗

Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscs.LG

A simple method reduces bias in LLM auto-evaluators by controlling output length.

problem Bias in LLM auto-annotators, particularly length bias.

method Simple regression analysis to control for length difference and other mediators.

result Improved robustness and increased correlation with human preferences.

bias reduction LLM evaluation length control regression analysis auto-annotators

2024-04-06abs ↗pdf ↗

Normalized Attention Without Probability Cagecs.LG

New attention model removes softmax, improving sequence length bias.

problem Softmax attention's limitations and sequence length bias.

method Replaces softmax with normalization in self-attention.

result Normalization leads to more robust and bias-free attention.

attention models softmax sequence length bias normalization

2020-05-19abs ↗pdf ↗

Ergodic Inference: Accelerate Convergence by Optimisationcs.LG

Improves MCMC and VI methods by optimising simulation bias.

problem Limitations of MCMC and VI methods in practice.

method Hybrid method combining optimisation with MCMC simulation.

result Produces low-biased samples with better balance between bias and efficiency.

MCMC VI Optimisation Simulation Bias Machine Learning

2018-05-25abs ↗pdf ↗

BSDAR: Beam Search Decoding with Attention Reward in Neural Keyphrase Generationcs.CL

Improved neural keyphrase generation by beam search with reward functions.

problem Sequence length bias and beam diversity issues in neural keyphrase generation.

method Beam search decoding strategy with word-level and ngram-level reward functions.

result Significant improvement in generating diverse and accurate keyphrases.

neural keyphrase generation beam search reward functions sequence length bias beam diversity

2019-09-17abs ↗pdf ↗

Rissanen Data Analysis: Examining Dataset Characteristics via Description Lengthcs.LG

Method estimates dataset utility via minimal program length proxy.

problem Determining if data labels are generated by useful subroutines.

method Rissanen Data Analysis (RDA) estimates minimum description length (MDL) as a proxy.

result Method reveals dataset characteristics and utility in various NLP settings.

data analysis description length NLP subroutine utility gender bias

2021-03-05abs ↗pdf ↗

Consistency of a Recurrent Language Model With Respect to Incomplete Decodingcs.LG

Recurrent models can produce infinite sequences, causing bias; new methods prevent this.

problem Inconsistency in decoding infinite-length sequences from recurrent language models.

method Defined and proved inconsistency of common decoding algorithms; proposed remedies.

result Proposed methods prevent inconsistency in practice.

recurrent models incomplete decoding inconsistency top-k sampling nucleus sampling

2020-02-06abs ↗pdf ↗

Anytime Monte Carlostat.CO

This work addresses real-time constraints in Monte Carlo simulations.

problem Real-time constraints in Monte Carlo simulations, especially in SMC algorithms.

method Proposes an anytime framework using a continuous-time Markov jump process to manage real-time computing budgets.

result Eliminates length bias in the final state's distribution of MCMC algorithms under real-time constraints.

Monte Carlo Real-time constraints Markov chain Monte Carlo Sequential Monte Carlo Length bias

2016-12-10abs ↗pdf ↗

Bias in the distribution of holonomy on compact hyperbolic 3-manifoldsmath.NT

The study reveals a persistent bias in the distribution of holonomy on compact hyperbolic 3-manifolds.

problem The distribution of holonomy on compact hyperbolic 3-manifolds is not uniformly distributed.

method An asymptotic count of closed geodesics by their length and holonomy, and analysis of spectral parameters.

result A normalized, smoothed bias count of holonomy is distributed according to a probability distribution, controlled by the number of zero spectral parameters.

holonomy compact hyperbolic 3-manifolds spectral parameters geodesics bias distribution

2022-06-29abs ↗pdf ↗

Understanding Simplicity Bias towards Compositional Mappings via Learning Dynamicscs.LG

The study reveals simplicity bias in neural networks leading to better compositional mappings.

problem Understanding when and how to encourage neural networks to learn compositional mappings.

method Examined compositional mappings through coding length and gradient descent dynamics.

result Neural networks tend to learn the simplest bijections, explaining their good generalization.

neural networks compositional mappings simplicity bias gradient descent Kolmogorov complexity

2024-09-15abs ↗pdf ↗

Accurate Evaluation of Quickest Changepoint Detectors via Non-parametric Survival Analysiscs.LG

Non-parametric estimators improve quickest changepoint detection under irregular sequence lengths.

problem Limited and irregular sequence lengths hinder application of ARL and ADD in QCD.

method Analogies with survival analysis to model detection probabilities under truncation.

result KM-ARL and KM-ADD non-parametric estimators are asymptotically unbiased.

quickest changepoint detection non-parametric survival analysis average run length average detection delay sequence truncation

2026-05-11abs ↗pdf ↗

Improved Path-length Regret Bounds for Banditscs.LG

Improved path-length regret bounds for adaptive and oblivious adversaries.

problem Adaptive and oblivious adversaries in multi-armed bandit and linear bandit problems.

method Developed two new algorithms based on optimistic mirror descent framework with novel techniques.

result Strictly improved path-length bounds for adaptive adversary and better results for oblivious adversary.

bandits regret optimistic mirror descent adaptive adversary oblivious adversary

2019-01-29abs ↗pdf ↗

Bias-Aware Conformal Prediction for Metric-Based Imaging Pipelinesstat.ML

Study improves confidence measures in medical imaging pipelines by addressing bias.

problem Bias in metric-based imaging pipelines compromises the efficiency of prediction intervals.

method Formalized symmetric and asymmetric CP formulations, analyzed bias effects, and validated empirically.

result Symmetric intervals are inflated by bias, while asymmetric intervals remain unaffected.

Bias Conformal Prediction Medical Imaging Confidence Intervals Radiation Therapy Planning

2024-10-07abs ↗pdf ↗

Provable Benefits of Sinusoidal Activation for Modular Additioncs.LG

Sine activation functions enable two-layer neural networks to learn modular addition more efficiently.

problem Learning modular addition with two-layer neural networks.

method Introduced and analyzed sine activation functions, providing theoretical and empirical evidence.

result Sine activation functions allow for constant-width network realizations of modular addition, whereas ReLU networks require linear width scaling.

modular addition two-layer neural networks sine activation expressivity gap generalization bound

2025-11-28abs ↗pdf ↗

Towards Understanding Inductive Bias in Transformers: A View From Infinitycs.LG

Transformers tend to learn more symmetric functions in sequence data.

problem Understanding inductive bias in Transformers with infinitely over-parameterized models.

method Analyzing Transformers in the Gaussian process limit, using representation theory of the symmetric group.

result Transformers are biased towards more permutation symmetric functions, and this can be quantitatively predicted.

Transformers Inductive Bias Gaussian Process Symmetric Group Permutation Symmetry

2024-02-07abs ↗pdf ↗

Towards Learning Convolutions from Scratchcs.LG

New method learns convolution-like structures from scratch.

problem Learning convolution-like structures from data.

method Minimum description length principle and

β

-LASSO algorithm.

result Learned architectures achieve state-of-the-art accuracies.

Convolution Machine Learning Architecture Search Image Classification CIFAR-10

2020-07-27abs ↗pdf ↗

Decoupled Q-Chunkingcs.LG

Decouples critic chunk length from policy to improve policy reactivity and performance.

problem Bootstrapping bias and difficulty in extracting optimal policies from chunked critics.

method Optimizes policy against a distilled critic for partial action chunks, allowing shorter chunks for policy.

result Reliably outperforms prior methods on long-horizon offline goal-conditioned tasks.

Temporal-Difference Value Backup Policy Extraction Chunked Critics Bootstrapping Bias

2025-12-11abs ↗pdf ↗

XEM: An Explainable-by-Design Ensemble Method for Multivariate Time Series Classificationcs.LG

XEM improves multivariate time series classification with explainable models.

problem Multivariate time series classification challenges.

method Hybrid ensemble method combining explicit boosting-bagging and implicit divide-and-conquer.

result XEM outperforms state-of-the-art MTS classifiers on public datasets.

time series classification ensemble methods explainable models bias-variance trade-off hybrid approach

2020-05-07abs ↗pdf ↗

A Random Matrix Perspective of Echo State Networks: From Precise Bias--Variance Characterization to Optimal Regularizationstat.ML

New analysis reveals optimal regularization for ESNs, avoiding double descent.

problem Characterizing and optimizing Echo State Networks (ESNs) for precise bias-variance.

method Random matrix theory applied to ESNs in a teacher-student setting.

result ESNs achieve lower MSE with limited training samples and teacher memory.

Echo State Networks Random Matrix Theory Regularization Bias-Variance Double Descent

2025-09-26abs ↗pdf ↗

Temporal and Object Quantification Networkscs.LG

TOQ-Nets learn to recognize complex temporal events with varying objects and sequences.

problem Recognizing complex relational-temporal events with varying numbers of objects and sequence lengths.

method Neuro-symbolic networks with reasoning layers for finite-domain quantification over objects and time.

result TOQ-Nets can generalize to scenarios with more objects than training data and temporal warpings.

Temporal events Quantification networks Complex patterns Varying objects Sequence lengths

2021-06-10abs ↗pdf ↗

Quantifying Confounding Bias in Neuroimaging Datasets with Causal Inferencecs.LG

New method quantifies biases in neuroimaging datasets.

problem Pooling neuroimaging datasets can introduce biases.

method Causal inference to quantify confounding factors.

result Simple pooling often leads to biased training data.

neuroimaging bias causal inference confounding MRI

2019-07-09abs ↗pdf ↗

Contextual Online Pricing with (Biased) Offline Datacs.LG

Study contextual online pricing with biased offline data, achieving optimal regret bounds.

problem Contextual online pricing with biased offline data.

method Identify

δ^2

to measure data bias, use OFU policy and robust variant for unknown bias.

result Achieve minimax-optimal regret bounds for contextual pricing.

contextual pricing online learning offline data regret bounds bias mitigation

2025-07-03abs ↗pdf ↗

Technical Analysis and Discrete False Discovery Rate: Evidence from MSCI Indicesq-fin.ST

Study evaluates technical trading rules on various markets, introduces DFRD+/- method.

problem Evaluating the profitability and robustness of technical trading rules across different markets.

method Investigated 21,000 technical trading rules on 12 markets over 12 years, introduced DFRD+/- method.

result DFRD+/- method is adaptive and more powerful, accommodating discrete p-values.

technical analysis false discovery rate financial markets rolling forward structures dynamic portfolios

2018-11-16abs ↗pdf ↗

Mixture of Step Returns in Bootstrapped DQNcs.LG

MB-DQN uses different backup lengths for improved reinforcement learning.

problem Improving reinforcement learning with multi-step returns.

method Integrates multi-step returns into bootstrapped DQN with different backup lengths.

result MB-DQN maintains advantages of different backup lengths.

reinforcement learning bootstrapped DQN multi-step returns TD-lambda Atari 2600

2020-07-16abs ↗pdf ↗

UCB Momentum Q-learning: Correcting the bias without forgettingstat.ML

UCBMQ improves Q-learning by adding momentum to correct bias and limit regret.

problem Improving Q-learning's bias and regret in reinforcement learning.

method UCBMQ combines Q-learning with an upper confidence bound and momentum term.

result UCBMQ guarantees a regret of

O(\sqrt{H^3SAT}+ H^4 S A )

with a linear second-order term in

S

reinforcement learning Q-learning momentum upper confidence bound regret

2021-03-01abs ↗pdf ↗

Fairness in Forecasting and Learning Linear Dynamical Systemscs.LG

The paper tackles fairness in forecasting and learning linear dynamical systems.

problem Under-representation bias in training data for multiple subgroups.

method Introducing subgroup-fair and instant-fair learning of LDS from multiple trajectories of varying lengths, using hierarchies of convexifications of non-commutative polynomial optimisation problems.

result Empirical results show both the beneficial impact of fairness considerations on statistical performance and encouraging effects of exploiting sparsity on run time.

fairness forecasting linear dynamical systems subgroup-fair instant-fair

2020-06-12abs ↗pdf ↗

Learning to Encode Position for Transformer with Continuous Dynamical Modelcs.LG

New method for Transformer models to encode position information without sequential bias.

problem Lack of flexible and learnable position encoding for Transformer models.

method Continuous dynamical model to learn position encoding.

result Consistent improvements over baselines in various NLP tasks.

Transformer Position Encoding Continuous Dynamical Model Neural ODE Machine Translation

2020-03-13abs ↗pdf ↗

Finite sample properties of power-law cross-correlations estimatorsphysics.data-an

We study finite sample properties of estimators of power-law cross-correlations -- detrended cross-correlation analysis (DCCA), height cross-correlation analysis (HXA) and detrending moving-average cross-correlation analysis (DMCA) -- with a special focus on short-term memory bias as well as power-law coherency. Presen…

2014-09-24abs ↗pdf ↗

Learning GFlowNets from partial episodes for improved convergence and stabilitycs.LG

New method trains GFlowNets from partial episodes to improve convergence and stability.

problem Improving convergence and stability of GFlowNets training.

method Introducing SubTB(

λ

) for GFlowNet training from partial action subsequences.

result SubTB(

λ

) accelerates GFlowNet convergence and enables training in longer action sequences.

GFlowNets partial episodes subtrajectory balance reinforcement learning convergence

2022-09-26abs ↗pdf ↗

Improving Sequential Determinantal Point Processes for Supervised Video Summarizationcs.LG

This paper improves video summarization using a new algorithm and dataset.

problem Efficiently summarizing videos for browsing and searching.

method Improves sequential determinantal point process (SeqDPP) with a large-margin algorithm and a new probabilistic distribution.

result Significantly improved video summarization model with better user input integration and diversity.

video summarization sequential determinantal point process supervised learning user input exposure bias

2018-07-28abs ↗pdf ↗

Hidden Synergy:

L_1

Weight Normalization and 1-Path-Norm Regularizationcs.LG

PSiLON Net uses $L_1$ weight normalization and 1-path-norm regularization for efficient learning and sparsity.

problem Efficient learning and sparsity in neural networks with limited data.

method PSiLON Net employs

L_1

weight normalization and 1-path-norm regularization to simplify the 1-path-norm and achieve efficient learning and near-sparse parameters.

result PSiLON Net achieves reliable optimization and strong performance in the small data regime.

L1 weight normalization 1-path-norm regularization efficient learning sparsity neural networks

2024-04-29abs ↗pdf ↗

Network reconstruction via the minimum description length principlestat.ML

A new method avoids overfitting in network reconstruction by using the minimum description length principle.

problem Determining the optimal model complexity in network reconstruction to prevent overfitting.

method Hierarchical Bayesian inference and weight quantization based on the minimum description length principle.

result The method yields increased accuracy in reconstructing both artificial and empirical networks.

network reconstruction minimum description length sparsity overfitting microbial communities

2024-05-02abs ↗pdf ↗

Enhancing Transformer-Based Foundation Models for Time Series Forecasting via Bagging, Boosting and Statistical Ensemblescs.LG

Enhanced TSFMs improve time series forecasting accuracy and reliability.

problem Variance, bias, and uncertainty in TSFMs' predictions on real data.

method Statistical and ensemble techniques including bagging, stacking, residual modeling, and prediction intervals.

result Hybrid models consistently outperform standalone TSFMs across multiple horizons.

time series forecasting foundation models ensemble methods statistical reasoning prediction intervals

2025-08-18abs ↗pdf ↗

Learning from a Biased Samplestat.ME

Proposes methods to learn from biased samples, ensuring robust decision rules.

problem Learning from biased samples can lead to poor performance in real-world applications.

method Modeling sampling bias, using distributionally robust optimization and deep learning.

result Proposes a method to minimize worst-case risk under various test distributions.

sampling bias empirical risk minimization distributionally robust optimization deep learning statistical guarantees

2022-09-05abs ↗pdf ↗

Policy Optimization Achieves Data-Dependent Regret Bounds in MDPs with Unknown Transitionscs.LG

New algorithm achieves data-dependent regret bounds in MDPs with unknown transitions.

problem Achieving best-of-both-worlds guarantees with data-dependent regret bounds in MDPs with unknown transitions.

method Optimistic follow-the-regularized-leader algorithm with new optimistic Q-function estimators and transition bonus.

result First-order, second-order, and path-length bounds with polylog(T) regret in the stochastic regime.

policy optimization Markov decision processes unknown transitions regret bounds optimistic algorithms

2026-06-30abs ↗pdf ↗

FinAI-BERT: A Transformer-Based Model for Sentence-Level Detection of AI Disclosures in Financial Reportsq-fin.CP

FinAI-BERT classifies AI disclosures in financial reports with high accuracy.

problem Systematic detection of AI-related disclosures in financial reports.

method Fine-tuned transformer-based model on a curated dataset.

result Achieved near-perfect classification performance (99.37% accuracy).

Transformer Financial NLP AI Disclosures Sentence-Level Classification Financial Reports

2025-06-29abs ↗pdf ↗

Underestimation Bias and Underfitting in Machine Learningcs.LG

Research shows bias in machine learning can be due to algorithmic flaws, not just data.

problem Underestimation bias in machine learning algorithms.

method Initial research to understand factors contributing to bias in classification algorithms.

result Regularization methods to address overfitting can also accentuate bias.

machine learning bias underfitting algorithmic flaws overfitting

2020-05-18abs ↗pdf ↗

Multi-period Learning for Financial Time Series Forecastingq-fin.ST

Enhances financial time series forecasting with a multi-period learning framework.

problem Accurate financial time series forecasting requires considering both short-term and long-term trends.

method Proposes a Multi-period Learning Framework (MLF) with three modules: Inter-period Redundancy Filtering, Learnable Weighted-average Integration, and Multi-period self-Adaptive Patching.

result Improves financial time series forecasting accuracy and efficiency.

financial time series forecasting multi-period learning self-attention modeling redundancy filtering weighted-average integration

2025-11-07abs ↗pdf ↗

A Benchmark Study of Machine Learning Models for Online Fake News Detectioncs.CL

Study benchmarks machine learning models for fake news detection.

problem Dataset bias and performance of fake news detection models.

method Benchmarked different machine learning models on three datasets.

result BERT and similar pre-trained models perform best for fake news detection.

fake news machine learning benchmark study BERT dataset bias

2019-05-12abs ↗pdf ↗

Trisections with Kirby-Thompson length 2math.GT

This paper proves a conjecture about trisections with a specific length.

problem Proving a conjecture about trisections with a specific length.

method Examining trisections with Kirby-Thompson length 2 and proving the conjecture.

result Proves the conjecture about length 2 trisection being a 4-manifold with length 0.

trisection length 2 Kirby-Thompson 4-manifold conjecture

2021-12-19abs ↗pdf ↗

Measuring Tail Risksq-fin.RM

Proposes a new tail risk measure based on the most probable maximum risk event size.

problem Current risk measures like VaR and ES are limited in their applicability and require specifying a confidence level.

method Develops a new risk measure called MPMR that does not require a confidence level and scales with the length of the time interval.

result The new risk measure, MPMR, scales with the number of observations by a power law, allowing for reliable estimations of long-term risks based on short-term estimations.

tail risk MPMR value at risk expected shortfall power law

2022-09-15abs ↗pdf ↗

Feature-Wise Bias Amplificationcs.LG

New feature selection methods reduce bias amplification in models.

problem Bias amplification in machine learning models.

method Analysis and development of feature selection algorithms.

result Feature-wise bias amplification can be mitigated through targeted feature selection.

bias amplification feature selection gradient descent machine learning convolutional neural networks

2018-12-21abs ↗pdf ↗

Power-law cross-correlations estimation under heavy tailsq-fin.ST

We examine the performance of six estimators of the power-law cross-correlations -- the detrended cross-correlation analysis, the detrending moving-average cross-correlation analysis, the height cross-correlation analysis, the averaged periodogram estimator, the cross-periodogram estimator and the local cross-Whittle e…

2016-02-17abs ↗pdf ↗

Evaluating LLMs in Finance Requires Explicit Bias Considerationcs.LG

Financial LLMs need explicit bias consideration to avoid invalid results.

problem Finance-specific biases inflate performance and contaminate backtests.

method Identified five recurring biases and proposed a Structural Validity Framework.

result Explicit bias consideration is necessary for valid deployment claims.

LLMs finance bias evaluation validity

2026-02-15abs ↗pdf ↗

Ground Truth Bias in External Cluster Validity Indicesstat.ML

It has been noticed that some external CVIs exhibit a preferential bias towards a larger or smaller number of clusters which is monotonic (directly or inversely) in the number of clusters in candidate partitions. This type of bias is caused by the functional form of the CVI model. For example, the popular Rand index (R…

2016-06-17abs ↗pdf ↗

Characterizing Bias in Post-Bandit Inference under Index Algorithmscs.LG

We analyze bias in post-bandit inference for stable index algorithms.

problem Bias in post-bandit inference for stable index algorithms.

method Empirical fluid approximation of sampling dynamics.

result Sharp leading-order expressions for bias and expected Z-statistic.

bias post-bandit inference index algorithms empirical fluid approximation

2026-08-02abs ↗pdf ↗

Mitigating Gender Bias Amplification in Distribution by Posterior Regularizationcs.CL

Mitigates gender bias amplification in model predictions.

problem Gender bias amplification in model predictions.

method Posterior regularization to mitigate bias.

result Almost removes gender bias amplification in model predictions.

gender bias natural language processing distribution posterior regularization model predictions

2020-05-13abs ↗pdf ↗

Filling Length in Finitely Presentable Groupsmath.GR

Filling length measures the length of the contracting closed loops in a null-homotopy. The filling length function of Gromov for a finitely presented group measures the filling length as a function of length of edge-loops in the Cayley 2-complex. We give a bound on the filling length function in terms of the log of an …

2000-08-03abs ↗pdf ↗