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.

168,657 papers · 148 categories

Trend · papers per month

4385128170 · May 202619922001200920172026
48 results for Trajectory Alignment

Cohen et al. (2021) show GD trajectories align on a bifurcation diagram.

problem Understanding the Edge of Stability (EoS) phenomenon in gradient descent.
method Empirical studies and rigorous mathematical proofs for two-layer networks and single-neuron networks.
result GD trajectories align on a specific bifurcation diagram independent of initialization.

Gradient descent trains both layers of a ReLU network to fit a linear model.

problem Training dynamics of a ReLU network to fit a linear target function.
method Jointly training both layers of a one-hidden-layer ReLU network in a realizable setting with Gaussian inputs and labels.
result Gradient descent from a small random initialization converges to a global minimizer at a linear rate with optimal sample complexity.

SGD's training dynamics align with Hessian and gradient spectra in high-dimensional classification tasks.

problem Understanding the spectra of Hessian and gradient matrices in high-dimensional classification tasks.
method Rigorous analysis of SGD dynamics and spectra of Hessian and gradient matrices.
result SGD trajectory and emergent outlier eigenspaces align with a common low-dimensional subspace in multi-class high-dimensional mixtures and neural networks.

Framework improves clinical timeline reconstruction from text and tables.

problem Temporal precision and event timing in clinical narratives and EHRs.
method Retrieval-augmented multimodal alignment framework.
result Consistently improves absolute timestamp accuracy and temporal concordance.

The vision of automated driving is to increase both road safety and efficiency, while offering passengers a convenient travel experience. This requires that autonomous systems correctly estimate the current traffic scene and its likely evolution. In highway scenarios early recognition of cut-in maneuvers is essential f…

2019-07-25abs ↗pdf ↗

Signed-permutation coordinate transport improves model alignment across checkpoints.

problem Improper alignment of coordinate-indexed objects across model checkpoints.
method Introduces sign-marginalized Hungarian matching and coordinate-preserving transport.
result Recovering signed-permutation gauge improves coordinate alignment and model performance.

A new framework for averaging spatio-temporal signals using optimal transport and soft alignments.

problem Averaging complex datasets with time and spatial components.
method Inspired by DTW, OT, and UOT, a new loss function is proposed to address shifts in time, space, and population size.
result The proposed loss function can be used to compute spatio-temporal barycenters efficiently.

Diffusion models' sampling paths lie in a low-dimensional subspace, resembling boomerangs.

problem Understanding the geometric structure of diffusion-based generative models.
method Characterization of deterministic sampling trajectories using low-dimensional subspace and kernel-estimated data modeling.
result Sampling trajectories in diffusion models are confined to a low-dimensional subspace and exhibit a boomerang shape.

SARA uses similarity to learn rewards robustly and adaptively.

problem Robustness to labeler errors and adaptability to diverse feedback formats.
method Contrastive framework that learns latent representations and computes rewards as similarities.
result Strong performance on offline RL benchmarks and diverse applications.

TRACE improves conformal prediction for multi-dimensional outputs.

problem Challenges in constructing valid and informative conformal prediction regions for multi-dimensional outputs.
method TRACE uses transport alignment in diffusion and flow matching models to define nonconformity scores.
result TRACE yields valid and adaptive conformal prediction regions for multimodal and non-convex distributions.

Proposes a new method for dynamic treatment regimes that improves sample efficiency and stability.

problem Challenges in estimating optimal treatments for individuals with dynamic decision-making stages.
method Focuses on prioritizing alignment between observed and optimal treatment trajectories across decision stages.
result Improves sample efficiency and stability of IPWE-based methods by relaxing the alignment requirement.

A generative model with a disentangled representation allows for independent control over different aspects of the output. Learning disentangled representations has been a recent topic of great interest, but it remains poorly understood. We show that even for GANs that do not possess disentangled representations, one c…

2018-12-04abs ↗pdf ↗

SGD updates align with a low-rank subspace but do not lead to further loss reduction.

problem Understanding the training dynamics of deep neural networks, particularly the role of the dominant subspace.
method Exploring whether neural networks can be trained within the dominant subspace of the loss Hessian.
result SGD updates, when projected onto the dominant subspace, do not decrease the training loss further, suggesting spurious alignment.

Identifies bilinear systems from a single trajectory with optimal sample complexity.

problem Learning bilinear systems from a single trajectory of states and inputs.
method Uses a mild marginal mean-square stability assumption and martingale small-ball condition.
result Sample complexity and statistical error rates are optimal.

Paper examines adversarial attacks on weather forecasting models, focusing on TC trajectory prediction.

problem Adversarial attacks can mislead downstream TC trajectory predictions in DLWF models.
method Proposes Cyc-Attack, a method using a surrogate model and skewness-aware loss function to generate adversarial TC paths.
result Cyc-Attack achieves higher true positive rates and lower false alarm rates compared to conventional methods.

Paper proposes a faster, higher-quality RL method for text-to-image models.

problem Improving image quality and prompt alignment in RL post-training of text-to-image models.
method Online RL variant that reduces variance by sampling paired trajectories and optimizing flow velocity.
result Method converges faster and yields higher output quality and prompt alignment.

MFMs enable efficient reward alignment for generative models.

problem Computational bottleneck in controlling generative models.
method Meta Flow Maps (MFMs) extend consistency models and flow maps to stochastic regime for efficient value function estimation.
result MFMs enable inference-time steering and unbiased, off-policy fine-tuning to general rewards efficiently.

Enhances RL in target domains with limited data using augmented return.

problem Utilize data from an accessible source domain to improve policy learning in a target domain with scarce data.
method Return Augmented Decision Transformer (REAG) method, which augments the return in the source domain to align with the target domain's optimal trajectory distribution.
result The proposed REAG method achieves the same level of suboptimality as without a dynamics shift, enhancing DT type frameworks' performance in off-dynamics RL.

Paper proposes a probabilistic alignment method for domain adaptation.

problem Latent distribution mismatch and miscalibrated uncertainty in adapting large-scale models.
method Bayesian latent transport framework with PAC-Bayesian regularization.
result Reduction in latent manifold discrepancy and improved uncertainty calibration.

DeepFRC learns both alignment and classification of functional data in one model.

problem Phase variability in functional data obscures underlying patterns and degrades model performance.
method End-to-end deep learning framework that combines diffeomorphic warping functions and a classifier.
result DeepFRC outperforms state-of-the-art methods in both alignment quality and classification accuracy.

GWIL uses Gromov-Wasserstein distance to align expert and imitation agent states.

problem Cross-domain imitation learning challenges due to different system dimensions and stationary distributions.
method Gromov-Wasserstein Imitation Learning (GWIL) using Gromov-Wasserstein distance.
result GWIL effectively aligns expert and imitation agent states in various continuous control domains.

GMC benchmark isolates retrieval in Transformers, revealing max-margin alignment.

problem Understanding how Transformers develop match-and-copy behavior on natural data.
method Introducing Gaussian Match-and-Copy (GMC) as a minimalist benchmark.
result Gradient descent drives parameters to diverge while aligning with max-margin separator.

Paper establishes a generalization bound for gradient flow using a data-dependent kernel.

problem Understanding the generalization properties of gradient-based optimization methods.
method Establishes a generalization bound for gradient flow through a data-dependent kernel called the loss path kernel (LPK).
result The LPK captures the entire training trajectory and leads to tighter generalization guarantees.

Gradient descent with small random init mimics spectral methods for low-rank matrix recovery.

problem Reconstructing a low-rank matrix from few measurements.
method Gradient descent with small random initialization followed by a few iterations.
result Gradient descent from small random init converges to a well-generalizing solution.

Deep model improves option pricing for CSI 300 index with sentiment and volatility features.

problem Challenges in real market option pricing, especially with constant volatility assumption.
method Deep Forward-Backward Stochastic Differential Equation (FBSDE) framework with dual-network architecture.
result Significant reduction in MAE and MAPE compared to BSM model.

AlignFlow improves FGMs by optimizing noise and data alignment.

problem Optimal Transport methods for FGMs are limited by scalability issues.
method Introduces Semi-Discrete Optimal Transport (SDOT) to enhance FGM training.
result AlignFlow scales well to large datasets and model architectures.

NeuralMD accelerates protein-ligand binding simulations 1Kx faster.

problem Accurate and efficient simulation of protein-ligand binding dynamics.
method Physics-informed multi-grained group symmetric framework with BindingNet and augmented neural differential equation solver.
result Achieves over 1Kx speedup and up to 15x reduction in reconstruction error compared to standard methods.

Auto-CEI improves LLM reasoning by balancing assertiveness and conservativeness.

problem Hallucinations and laziness in LLM reasoning tasks.
method Expert Iteration explores reasoning trajectories, guiding incorrect paths back on track and promoting appropriate 'I don't know' responses.
result Auto-CEI achieves superior alignment in logical reasoning, mathematics, and planning tasks.

Paper introduces OTR for efficient offline RL in surgical robotics.

problem Lack of annotated datasets for offline RL in surgical robotics.
method OTR algorithm using Optimal Transport to assign rewards to unlabeled trajectories.
result OTR enables efficient policy learning from large datasets without handcrafted rewards.

CB-RL solves complex decision-making problems with contextual information and exogenous events.

problem Optimal policy in strategic decision-making problems that depend on environmental configuration and exogenous events.
method Contextual Bilevel Reinforcement Learning (CB-RL) with a stochastic Hyper Policy Gradient Descent (HPGD) algorithm.
result Demonstrated convergence and performance of the HPGD algorithm for reward shaping and tax design.

SDM Policy accelerates inference for robotic tasks while maintaining high action quality.

problem Prolonged inference times in diffusion-based policies for high-frequency control tasks.
method Two-stage optimization: score matching and distribution matching; dual-teacher mechanism.
result 6x inference speedup with state-of-the-art action quality.

New protocol evaluates synthetic data for temporal consistency.

problem Synthetic data generators can produce invalid timestamps and trajectories.
method Characterize datasets by four properties, then measure timestamp validity and dynamics.
result Temporal fidelity must be measured, not inferred from static data.

Study SGD dynamics in sequence models, revealing training phases and influence of sequence length.

problem Understanding SGD in sequence models like attention networks.
method Derived closed-form population loss and analyzed SGD dynamics for SSI models.
result Two distinct training phases: escape from uninformative initialization and alignment with target subspace.

Calibrated PRMs improve inference efficiency for LLMs by dynamically adjusting compute budgets.

problem Poor calibration of PRMs leads to overestimation of success probabilities in partial reasoning steps.
method Quantile regression for calibration, instance-adaptive scaling (IAS) framework.
result Calibrated PRMs reduce inference costs while maintaining accuracy, especially on confident problems.

DHLNN improves deep hedging for financial derivatives with faster convergence and better stability.

problem Challenges in computational inefficiency, sensitivity to noisy data, and optimization complexity in deep hedging methods.
method Integrates periodic fixed-gradient optimization and linearized training dynamics to stabilize and accelerate deep learning model training.
result Demonstrates faster convergence, improved stability, and superior hedging performance across diverse market scenarios.

We link disjoint longitudinal data for rare disease patients using latent representations and mixed-effects regression.

problem Analyzing treatment switches in rare diseases with limited data and changing measurement instruments.
method We embed item values into a shared latent space using variational autoencoders and apply mixed-effects regression to quantify treatment effects.
result Our approach allows for statistical inference and quantifies the impact of treatment switches in spinal muscular atrophy.

Unified LP framework for offline reward learning from human demonstrations and feedback.

problem Reward learning from human demonstrations and feedback with robustness and sample efficiency.
method A novel linear programming framework for offline reward learning.
result Unified LP framework achieves better performance compared to MLE.