The paper examines how spike strengths and alignments affect overfitting in linear regression models.
problem The impact of spike strengths and alignments on overfitting in linear regression models.
method Characterization of generalization error through exact expressions and analysis of spike strengths, aspect ratio, and target alignment.
result Increasing spike strength can lead to catastrophic overfitting before benign overfitting, especially in well-specified aligned problems.
The paper addresses rigid alignment of noisy patches, providing a polynomial time algorithm and convergence conditions.
problem Finding a rigid alignment of overlapping local views (patches) that minimizes alignment error in a noisy setting.
method Characterizes non-degeneracy based on kernel and positivity of a matrix, provides polynomial time algorithm for testing non-degeneracy, and uses Riemannian gradient descent for alignment.
result The algorithm converges locally linearly to a non-degenerate perfect alignment under certain conditions.
The paper addresses poor calibration in fine-tuned LLMs after preference alignment.
problem Poor calibration in fine-tuned Large Language Models (LLMs) after preference alignment.
method Proposes a calibration-aware fine-tuning approach to restore calibration without compromising model performance.
result Demonstrates the effectiveness of the proposed methods through extensive experiments.
New method trains deep neural networks without backpropagation.
problem Training deep neural networks efficiently and without backpropagation.
method Direct feedback alignment using random weights for error propagation.
result Achieves zero training error in convolutional and deep networks.
Study evaluates manifold alignment methods for noisy double pendulum dynamics.
problem Aligning manifolds of double pendulum dynamics under noise.
method Compared four manifold alignment methods: semi-supervised feature-level global and local.
result Local alignment methods were more robust to noise and faster.
New approach improves domain adaptation by relaxing distribution alignment constraints.
problem Improving domain adaptation when target distribution differs from source distribution.
method Asymmetrically-relaxed distribution alignment to minimize target error under varying conditions.
result Empirical and theoretical benefits demonstrated on synthetic and real datasets.
A new method reduces preference distortion in LLM alignment.
problem Vulnerability of traditional LLM alignment methods to human preference heterogeneity.
method Sign Estimator: A simple, provably consistent, and efficient estimator using binary classification loss.
result Substantially reduces preference distortion over a panel of simulated personas.
Proposes a regularization method for unsupervised domain adaptation that aligns predictions with target data's top singular vectors.
problem Domain adaptation challenges in high joint error scenarios.
method Regularizes classifier to align with unsupervised target data guided by label alignment property (LAP).
result The method improves performance in MNIST-USPS domain adaptation and cross-lingual sentiment analysis.
LLMs can fail to maximize aligned values even after training, due to irrational reasoning.
problem Value misalignment in LLMs' reasoning despite training alignment.
method Formalized rational value risk and decomposed estimation error.
result Rational value risk is widespread and cannot be fully eliminated.
Direct feedback alignment reduces data movement in neural networks.
problem Efficiency and energy-efficiency in training large neural networks.
method Sparse feedback matrix for local learning, reducing data movement and compute.
result Orders of magnitude improvement in data movement and 2x improvement in multiply-and-accumulate operations.
WiFi helps align and calibrate foot-mounted IMU trajectories.
problem Inertial drift and unknown initial states in FMIP.
method Graph-based SLAM with RSS measurements for WiFi APs.
result Aligns and calibrates trajectories accurately.
New method improves reliability of selecting individuals based on predicted treatment effects.
problem Reliability of selecting individuals based on predicted conditional average treatment effects (CATE) is unreliable.
method Denoised Conformal Alignment, combining proxy errors, variance estimation, and Benjamini-Hochberg selection.
result Significantly improved power in selecting individuals while maintaining false discovery rate control.
New algorithm shows neural networks can learn without full backpropagation.
problem Stochastic gradient descent with backpropagation is non-biologically plausible.
method Random and fixed backpropagation weights in a feedback alignment algorithm.
result Error converges to zero exponentially fast in overparameterized networks.
This research explores various sampling methods and probability distributions for hard alignment in sequence-to-sequence TTS synthesis.
problem Improving alignment accuracy in sequence-to-sequence text-to-speech synthesis.
method Investigated various sampling methods (greedy, beam, random) and probability distributions (Bernoulli, Concrete) for hard alignment.
result Deterministic search is more preferable than stochastic search for natural alignment transition.
Training-free source selection for LLM families with shared vocabularies
problem Source selection for LLM families with shared vocabularies
method Fisher alignment at vocabulary scale
result Fisher alignment is a cosine between kernel mean embeddings in the joint activation-error space
Decision-alignment evaluates uncertainty quantification for decision-relevant UQ
problem Evaluation of uncertainty quantification metrics
method Introduce decision-alignment
result Proper scoring rules align with decision utility
ADS explains object differences by quantifying and removing underlying properties.
problem Explaining differences between two object images.
method Align-Deform-Subtract (ADS) framework that uses semantic alignments and iterative quantification/removal of differences.
result ADS provides disentangled error measures explaining object differences in terms of underlying properties.
Alignment of neural network representations is influenced by SNR and sample size.
problem Understanding how neural network representations align across different conditions.
method Controlled training of neural networks on perturbed datasets, analyzing alignment and generalization.
result Alignment varies monotonically with SNR but non-monotonically with sample size, with minimal alignment near the interpolation threshold.
Transformer models align words through attention weights, closely approximating Optimal Transport.
problem Understanding the internal mechanism of transformer models in language processing.
method Empirical evidence and theoretical analysis of attention weights and their relation to Optimal Transport.
result Transformer models can simulate gradient descent on the dual of entropy-regularized OT problem, providing a theoretical foundation for token alignment.
Develops techniques to align word embeddings from different sources.
problem Aligning word embeddings from diverse datasets or methods.
method Simple closed-form techniques for optimal rotation, translation, and scaling.
result Maximizes cosine similarity and minimizes root mean squared errors.
Paper proposes new neural network learning algorithms inspired by predictive coding.
problem Finding biologically plausible alternatives to back-propagation of errors.
method Error-driven Local Representation Alignment (LRA-E) and Difference Target Propagation.
result Both proposed algorithms yield stable performance and strong generalization in training deeper, highly nonlinear networks.
The paper improves alignment methods for deep neural networks using geometric and spectral analysis.
problem Improving alignment methods for deep neural networks.
method Geometric and spectral analysis of residual Jacobian chains.
result Deterministic and margin-verified results on the transport of dominant singular subspaces across layers.
A new method for unsupervised domain adaptation using manifold learning.
problem Leveraging rich source domain information to target domain without labeled data.
method Discriminative Manifold Embedding and Alignment framework.
result Consistent transferability and discriminability achieved through manifold metric alignment.
Optimizes pruning masks for neural networks using probabilistic fine-tuning and PAC-Bayes bounds.
problem Improving neural network performance through adaptive pruning of weights.
method Optimizes stochastic pruning masks by minimizing expected loss, considering data-adaptive regularization and feature alignment.
result Probabilistic fine-tuning leads to improved test error over baseline methods in neural networks.
New training method improves neural network performance.
problem Improving neural network performance with direct feedback alignment.
method Direct feedback alignment with best practices and observations.
result Characterization of bottleneck effect in narrow layers.
Adversarial approach for optimizing multivariate performance measures in structured predictions.
problem Inconsistency between learner's objective and desired application performance.
method Adversarial training of structured predictions to optimize exact F-score or AER while matching training data properties.
result Improvement in multivariate performance measures for word alignment and named entity recognition.
New study reveals task alignment is key to ICL performance.
problem Understanding how task alignment affects generalization in in-context learning.
method Derived an exact expression for ICL generalization error in high dimensions under task covariance mismatch.
result Identified train-test task alignment as a key determinant of generalization in ICL.
Algorithm learns fair representations without sacrificing accuracy across groups.
problem Mitigating disparity among different demographic subgroups in classification.
method Balanced error rate and conditional alignment of representations.
result Improves utility-fairness trade-off on balanced datasets.
New method removes pseudo-label bias for unsupervised domain adaptation.
problem Class imbalance and distribution shift between domains.
method Implicit class-conditioned domain alignment without explicit pseudo-label optimization.
result Effective in handling within-domain class imbalance and between-domain class distribution shift.
Learning with transformed data using diffeomorphisms.
problem Training data transformation for improved prediction accuracy.
method Minimizes prediction error penalized by diffeomorphism distance to identity.
result Demonstrates potential of diffeomorphisms in large-dimensional estimation.
Improved deep learning for one-shot and open-set classification using alignment-based matching.
problem Limited data for one-shot classification and open-set recognition.
method Aligns images to reference images for classification, learns alignment mechanism.
result Significantly improved classification accuracy (e.g., 1.4% error rate in Omniglot, 46.5% in MiniImageNet).
New method prevents RLHF alignment collapse by accounting for policy's influence on reward model updates.
problem Iterative RLHF leads to alignment collapse where policies exploit RM's blind spots.
method Foresighted policy optimization (FPO) restores missing steering term via regularization.
result FPO prevents alignment collapse on LLM alignment pipelines using Llama-3.2-1B.
LLMs struggle with zero-shot annotation tasks due to model-internalized priors.
problem Impact of model-internalized priors on LLM performance in zero-shot annotation tasks.
method Investigated three dimensions: familiarity, decision stickiness, and susceptibility to misaligned task definitions.
result Nearly two-thirds of zero-shot errors are resistant to correction, with a rescue rate of 34.8%. Definition-Specific Familiarity (DSF) shows a positive association with model performance.
The paper explores how kernel eigenalignments affect generalization in KRR.
problem Achieving robust generalization in kernel methods.
method Direct connection between generalization and matrix eigenvectors/eigenvalues, focusing on finite-sample settings.
result Strong generalization requires increasing eigenvector alignment, eigenvalue magnitude, or gaps between eigenvalues.
This paper extends neural collapse to regression problems, revealing key features and structures.
problem Understanding the structure learned by deep neural networks in regression tasks.
method Established Neural Regression Collapse (NRC) across different models, analyzing feature and weight alignments.
result Deep neural regression models exhibit a collapsed feature space, aligning with target dimensions and covariances.
New method improves domain adaptation by aligning sub-domains.
problem Real-world domain adaptation with shifted class distributions.
method Rebalanced sub-domain alignment and novel generalization bound.
result Improves performance in shifted class distribution scenarios.
The study characterizes diffusion model generalization using data-dependent ridge manifolds.
problem Understanding where diffusion model-generated samples lie when not memorizing the training set.
method Introduced a time-dependent family of log-density ridge manifolds to characterize reverse-time inference.
result Generated samples evolve by a reach-align-slide mechanism, controlled by normal and tangential components of training error.
The paper analyzes the reward improvement of aligned policies in large language models.
problem Optimizing policies in large language models while staying close to a reference policy.
method Information-theoretic analysis and reduction to exponential order statistics.
result Information-theoretic upper bounds on reward improvement are derived.
Deep learning models show bias and variance are aligned, not in trade-off.
problem The classical bias-variance trade-off in deep learning models.
method Empirical evidence and theoretical analysis of bias and variance in deep learning models.
result Squared bias is approximately equal to variance for correctly classified sample points in deep learning models.
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.
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.
Paper resolves bias in ALFT training using generalized alignment games.
problem Systematic bias in estimating logarithmic rewards from small batches.
method Generalized Distributional Alignment Games, U-statistics, minimax polynomial estimators, Variance-Optimal Augmented Polynomial Optimization Program (AQP) Estimator.
result Proves optimal bias and accelerated convergence in ALFT training.
The paper studies and mitigates accuracy disparity in regression models.
problem Accuracy disparity between different demographic subgroups in high-stakes domains.
method Error decomposition theorem and distribution alignment algorithm.
result The proposed algorithm effectively mitigates accuracy disparity while maintaining predictive power.
This work explains early-stage alignment of logistic regression parameters with max-margin direction.
problem Early-stage alignment of logistic regression parameters with max-margin direction.
method Analyzes gradient descent dynamics and uses dataset geometry to track radial and tangential flows.
result Parameter vector weakly aligns with max-margin direction within O(exp(exp(-δ))) iterations, demonstrating faster weak alignment.
HSACC improves multi-view clustering of incomplete data.
problem Challenges in clustering incomplete multi-view data.
method Hierarchical Semantic Alignment and Cooperative Completion framework.
result HSACC outperforms state-of-the-art methods on benchmark datasets.
New learning rules for wide neural networks without backpropagation.
problem Training wide neural networks efficiently and without backpropagation.
method Input-weight alignment driven by gradient descent in the NTK regime.
result Biologically-motivated learning rules equivalent to backpropagation in wide networks.
Unified approach to domain generalization by aligning gradients and Hessians.
problem Developing models that generalize well across unseen domains.
method Moment Alignment, extending transfer measure to DG, aligning derivatives across domains.
result Moment Alignment unifies gradient and Hessian matching approaches, improving generalizability.
New research shows label refinement and weak training have limitations for aligning LLMs.
problem Limitations of refinement methods for aligning large language models.
method Analyzed probabilistic assumptions and alternative approaches to label refinement and weak training.
result Label refinement and weak training suffer from irreducible error, leaving a performance gap.