Search: “research”

Information-Theoretic Generalization Bounds for Stochastic Gradient Descent with Predictable Virtual Noise

arXiv:2605.00064v1 Announce Type: new Abstract: Information-theoretic generalization bounds analyze stochastic optimization by relating expected generalization error to the mutual information between

Research

Intrinsic Gradient Suppression for Label-Noise Prompt Tuning in Vision-Language Models

DGX agent

arXiv:2605.00591v1 Announce Type: new Abstract: Contrastive vision-language models like CLIP exhibit remarkable zero-shot generalization. However, prompt tuning remains highly sensitive to label noise

researcharxiv-cs-cv

4 May 2026

Is Textual Similarity Invariant under Machine Translation? Evidence Based on the Political Manifesto Corpus

arXiv:2605.00618v1 Announce Type: new Abstract: We investigate the extent to which cosine similarity between paragraph embeddings is invariant under machine translation, using the Manifesto Corpus of

It's Never Too Late: Noise Optimization for Collapse Recovery in Trained Diffusion Models

arXiv:2601.00090v2 Announce Type: replace Abstract: Contemporary text-to-image models exhibit a surprising degree of mode collapse, as can be seen when sampling several images given the same text prom

Knowing when to trust machine-learned interatomic potentials

arXiv:2605.00640v1 Announce Type: new Abstract: Prevailing machine-learned interatomic potential (MLIP) uncertainty-quantification methods rely on ensembles of independently trained backbones. These m

Koopman-Assisted Reinforcement Learning

arXiv:2403.02290v2 Announce Type: replace-cross Abstract: The Bellman equation and its continuous form, the Hamilton-Jacobi-Bellman equation, are ubiquitous in reinforcement learning and control theor

LandSegmenter: Towards a Flexible Foundation Model for Land Use and Land Cover Mapping

arXiv:2511.08156v2 Announce Type: replace Abstract: Land Use and Land Cover (LULC) mapping is a fundamental task in Earth Observation (EO). However, current LULC models are typically developed for a s

LASE: Language-Adversarial Speaker Encoding for Indic Cross-Script Identity Preservation

arXiv:2605.00777v1 Announce Type: cross Abstract: A speaker encoder used in multilingual voice cloning should treat the same speaker identically regardless of which script the audio was uttered in. Of

Last-Iterate Analyses of FTRL with the 1/2-Tsallis Entropy in Stochastic Bandits

arXiv:2510.22819v2 Announce Type: replace Abstract: The convergence analysis of online learning algorithms is central to machine learning theory, where the last-iterate convergence is particularly imp

Learning Multimodal Energy-Based Model with Multimodal Variational Auto-Encoder via MCMC Revision

arXiv:2605.00644v1 Announce Type: new Abstract: Energy-based models (EBMs) are a flexible class of deep generative models and are well-suited to capture complex dependencies in multimodal data. Howeve

Let ViT Speak: Generative Language-Image Pre-training

arXiv:2605.00809v1 Announce Type: new Abstract: In this paper, we present extbf{Gen}erative extbf{L}anguage-extbf{I}mage extbf{P}re-training (GenLIP), a minimalist generative pretraining framework for

Leveraging Vision-Language Models as Weak Annotators in Active Learning

arXiv:2605.00480v1 Announce Type: new Abstract: Active learning aims to reduce annotation cost by selectively querying informative samples for supervision under a limited labeling budget. In this work

LLM DNA: Tracing Model Evolution via Functional Representations

arXiv:2509.24496v3 Announce Type: replace Abstract: The explosive growth of large language models (LLMs) has created a vast but opaque landscape: millions of models exist, yet their evolutionary relat

Lost in State Space: Probing Frozen Mamba Representations

arXiv:2605.00253v1 Announce Type: new Abstract: Mamba's recurrent state h_t is, by construction, a compressed summary of every token seen so far. This raises a tempting hypothesis: if we extract token

Making Every Verified Token Count: Adaptive Verification for MoE Speculative Decoding

arXiv:2605.00342v1 Announce Type: new Abstract: Tree-based speculative decoding accelerates autoregressive generation by verifying multiple draft candidates in parallel, but this advantage weakens for

Matroid Algorithms Under Size-Sensitive Independence Oracles

arXiv:2605.00201v1 Announce Type: cross Abstract: The standard oracle model for matroid algorithms assumes that each independence query can be answered in constant time, regardless of the size of the

Mean-field limit from general mixtures of experts to quantum neural networks

arXiv:2501.14660v2 Announce Type: replace-cross Abstract: In this work, we study the asymptotic behavior of Mixture of Experts (MoE) trained via gradient flow on supervised learning problems. Our main

MMAudioReverbs: Video-Guided Acoustic Modeling for Dereverberation and Room Impulse Response Estimation

arXiv:2605.00431v1 Announce Type: cross Abstract: Although recent video-to-audio (V2A) models excelled at synthesizing semantically plausible sounds from visual inputs, they do not explicitly model ro

Modeling Subjective Urban Perception with Human Gaze

arXiv:2605.00764v1 Announce Type: new Abstract: Urban perception describes how people subjectively evaluate urban environments, shaping how cities are experienced and understood. Existing computationa

Near-optimal and Efficient First-Order Algorithm for Multi-Task Learning with Shared Linear Representation

arXiv:2605.00473v1 Announce Type: new Abstract: Multi-task learning (MTL) has emerged as a pivotal paradigm in machine learning by leveraging shared structures across multiple related tasks. Despite i

NRGPT: An Energy-based Alternative for GPT

arXiv:2512.16762v3 Announce Type: replace Abstract: Generative Pre-trained Transformer (GPT) architectures are the most popular design for language modeling. Energy-based modeling is a different parad

Observable Performance Does Not Fully Reflect System Organization: A Multi-Level Analysis of Gait Dynamics Under Occlusal Constraint

arXiv:2605.00778v1 Announce Type: new Abstract: In biomechanical systems, observable performance is often used as a proxy for underlying system organization. However, this assumption implicitly presum

Odysseus: Scaling VLMs to 100+ Turn Decision-Making in Games via Reinforcement Learning

arXiv:2605.00347v1 Announce Type: cross Abstract: Given the rapidly growing capabilities of vision-language models (VLMs), extending them to interactive decision-making tasks such as video games has e

On the Expressive Power of Contextual Relations in Transformers

arXiv:2603.25860v2 Announce Type: replace-cross Abstract: Transformer architectures have achieved remarkable empirical success in modeling contextual relations, yet a clear understanding of their expr

Optimal hypersurface decision trees

arXiv:2509.12057v3 Announce Type: replace Abstract: The study of optimal decision trees has gained increasing attention in recent years; however, despite substantial progress, it still suffers from tw

[P] QLoRA Fine-Tuning of Qwen2.5-1.5B for CEFR English Proficiency Classification (A1–C2) [P]

researchr-machinelearning

This post likely describes a machine learning project implementing QLoRA (Quantized Low-Rank Adaptation) fine-tuning on the Qwen2.5-1.5B model to classify English language proficiency levels according

PhysEdit: Physically-Consistent Region-Aware Image Editing via Adaptive Spatio-Temporal Reasoning

arXiv:2605.00707v1 Announce Type: new Abstract: Image editing instructions are heterogeneous: a color swap, an object insertion, and a physical-action edit all demand different spatial coverage and di

PhysiGen: Integrating Collision-Aware Physical Constraints for High-Fidelity Human-Human Interaction Generation

arXiv:2605.00517v1 Announce Type: new Abstract: Despite substantial progress in text-driven 3D human motion synthesis, generating realistic multi-person interaction sequences remains challenging. Nota

Possibilistic Predictive Uncertainty for Deep Learning

arXiv:2605.00600v1 Announce Type: cross Abstract: Deep neural networks achieve impressive results across diverse applications, yet their overconfidence on unseen inputs necessitates reliable epistemic

Posterior Augmented Flow Matching

arXiv:2605.00825v1 Announce Type: new Abstract: Flow matching (FM) trains a time-dependent vector field that transports samples from a simple prior to a complex data distribution. However, for high-di

Privacy Amplification in Differentially Private Zeroth-Order Optimization with Hidden States

arXiv:2506.00158v2 Announce Type: replace Abstract: Zeroth-order optimization has emerged as a promising approach for fine-tuning large language models under differential privacy (DP) and memory const

Quantum Gradient-Based Approach for Edge and Corner Detection Using Sobel Kernels

arXiv:2605.00744v1 Announce Type: new Abstract: Edge detection refers to identifying points in a digital image where intensity changes sharply, indicating object boundaries or structural features. Cor

Quantum Interval Bound Propagation for Certified Training of Quantum Neural Networks

arXiv:2605.00747v1 Announce Type: cross Abstract: Quantum machine learning is a promising field for efficiently learning features of a dataset to perform a specified task, such as classification. Inte

Randomized Subspace Nesterov Accelerated Gradient

arXiv:2605.00740v1 Announce Type: cross Abstract: Randomized-subspace methods reduce the cost of first-order optimization by using only low-dimensional projected-gradient information, a feature that i

RAT+: Train Dense, Infer Sparse -- Recurrence Augmented Attention for Dilated Inference

arXiv:2602.18196v3 Announce Type: replace Abstract: Structured dilated attention has an appealing inference-time efficiency knob: it reduces the FLOPs of attention and the KV cache size by a factor of

REALM: An RGB and Event Aligned Latent Manifold for Cross-Modal Perception

arXiv:2605.00271v1 Announce Type: new Abstract: Event cameras provide several unique advantages over standard frame-based sensors, including high temporal resolution, low latency, and robustness to ex

Representation in large language models

arXiv:2501.00885v2 Announce Type: replace Abstract: The extraordinary success of recent Large Language Models (LLMs) on a diverse array of tasks has led to an explosion of scientific and philosophical

Rethinking LLM Ensembling from the Perspective of Mixture Models

arXiv:2605.00419v1 Announce Type: cross Abstract: Model ensembling is a well-established technique for improving the performance of machine learning models. Conventionally, this involves averaging the

Revealing graph bandits for maximizing local influence

arXiv:2605.00489v1 Announce Type: new Abstract: We study a graph bandit setting where the objective of the learner is to detect the most influential node of a graph by requesting as little information

Reward Modeling from Natural Language Human Feedback

arXiv:2601.07349v3 Announce Type: replace Abstract: Reinforcement Learning with Verifiable reward (RLVR) on preference data has become the mainstream approach for training Generative Reward Models (GR

Riemannian MeanFlow

arXiv:2602.07744v3 Announce Type: replace Abstract: Diffusion and flow models have become the dominant paradigm for generative modeling on Riemannian manifolds, with successful applications in protein

ROSA: Robust and Energy-Efficient Microring-Based Optical Neural Networks via Optical Shift-and-Add and Layer-Wise Hybrid Mapping

arXiv:2605.00032v1 Announce Type: cross Abstract: This work presents ROSA, a microring-based optical neural network architecture that improves robustness and energy efficiency using an optical shift-a

Scale-Aware Adversarial Analysis: A Diagnostic for Generative AI in Multiscale Complex Systems

arXiv:2605.00510v1 Announce Type: cross Abstract: Complex physical systems, from supersonic turbulence to the macroscopic structure of the universe, are governed by continuous multiscale dynamics. Whi

Scaling Reasoning Hop Exposes Weaknesses: Demystifying and Improving Hop Generalization in Large Language Models

arXiv:2601.21214v2 Announce Type: replace Abstract: Chain-of-thought (CoT) reasoning has become the standard paradigm for enabling Large Language Models (LLMs) to solve complex problems. However, rece

Score-based Greedy Search for Structure Identification of Partially Observed Linear Causal Models

arXiv:2510.04378v2 Announce Type: replace Abstract: Identifying the structure of a partially observed causal system is essential to various scientific fields. Recent advances have focused on constrain

Short Chains, Deep Thoughts: Balancing Reasoning Efficiency and Intra-Segment Capability via Split-Merge Optimization

arXiv:2602.03141v3 Announce Type: replace Abstract: While Large Reasoning Models (LRMs) have demonstrated impressive capabilities in solving complex tasks through the generation of long reasoning chai

SketchGuard: Scaling Byzantine-Robust Decentralized Federated Learning via Sketch-Based Screening

arXiv:2510.07922v4 Announce Type: replace Abstract: Decentralized Federated Learning (DFL) enables privacy-preserving collaborative training without centralized servers but remains vulnerable to Byzan

Smart Ensemble Learning Framework for Predicting Groundwater Heavy Metal Pollution

arXiv:2605.00056v1 Announce Type: new Abstract: Groundwater in the Densu Basin is increasingly threatened by heavy metal contamination, but conventional methods fail to capture the statistical complex