Search: “models”

Demystifying the Silence of Correctness Bugs in PyTorch Compiler

arXiv:2604.08720v1 Announce Type: cross Abstract: Performance optimization of AI infrastructure is key to the fast adoption of large language models (LLMs). The PyTorch compiler (torch.compile), a cor

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

DRBENCHER: Can Your Agent Identify the Entity, Retrieve Its Properties and Do the Math?

DGX agent

arXiv:2604.09251v1 Announce Type: new Abstract: Deep research agents increasingly interleave web browsing with multi-step computation, yet existing benchmarks evaluate these capabilities in isolation,

model-releasesarxiv-cs-ai

13 Apr 2026

Drift-Aware Online Dynamic Learning for Nonstationary Multivariate Time Series: Application to Sintering Quality Prediction

arXiv:2604.09358v1 Announce Type: new Abstract: Accurate prediction of nonstationary multivariate time series remains a critical challenge in complex industrial systems such as iron ore sintering. In

Dynamic sparsity in tree-structured feed-forward layers at scale

arXiv:2604.08565v1 Announce Type: cross Abstract: At typical context lengths, the feed-forward MLP block accounts for a large share of a transformer's compute budget, motivating sparse alternatives to

Applications

Efficient Spatial-Temporal Focal Adapter with SSM for Temporal Action Detection

arXiv:2604.09164v1 Announce Type: new Abstract: Temporal human action detection aims to identify and localize action segments within untrimmed videos, serving as a pivotal task in video understanding.

applicationsarxiv-cs-cv

EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers

arXiv:2604.09130v1 Announce Type: cross Abstract: As SE(3)-equivariant graph neural networks mature as a core tool for 3D atomistic modeling, improving their efficiency, expressivity, and physical con

Exploring Cross-lingual Latent Transplantation: Mutual Opportunities and Open Challenges

arXiv:2412.12686v3 Announce Type: replace Abstract: Current large language models (LLMs) often exhibit imbalances in multilingual capabilities and cultural adaptability, largely attributed to their En

researcharxiv-cs-cl

Agents

Fast-dVLM: Efficient Block-Diffusion VLM via Direct Conversion from Autoregressive VLM

arXiv:2604.06832v2 Announce Type: replace Abstract: Vision-language models (VLMs) predominantly rely on autoregressive decoding, which generates tokens one at a time and fundamentally limits inference

agentsarxiv-cs-cl

FIT-GNN: Faster Inference Time for GNNs that 'FIT' in Memory Using Coarsening

arXiv:2410.15001v5 Announce Type: replace Abstract: Scalability of Graph Neural Networks (GNNs) remains a significant challenge. To tackle this, methods like coarsening, condensation, and computation

FP8-RL: A Practical and Stable Low-Precision Stack for LLM Reinforcement Learning

arXiv:2601.18150v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) for large language models (LLMs) is increasingly bottlenecked by rollout (generation), where long output sequence

safetyarxiv-cs-cl

Generalization and Scaling Laws for Mixture-of-Experts Transformers

arXiv:2604.09175v1 Announce Type: cross Abstract: We develop a theory of generalization and scaling for Mixture-of-Experts (MoE) Transformers that cleanly separates active per-input capacity from rout

Applications

Identification and Anonymization of Named Entities in Unstructured Information Sources for Use in Social Engineering Detection

arXiv:2604.09016v1 Announce Type: cross Abstract: This study addresses the challenge of creating datasets for cybercrime analysis while complying with the requirements of regulations such as the Gener

applicationsarxiv-cs-ai

Integrated electro-optic attention nonlinearities for transformers

arXiv:2604.09512v1 Announce Type: new Abstract: Transformers have emerged as the dominant neural-network architecture, achieving state-of-the-art performance in language processing and computer vision

Tutorials

Learning General Representation of 12-Lead Electrocardiogram with a Joint-Embedding Predictive Architecture

arXiv:2410.08559v5 Announce Type: replace-cross Abstract: Electrocardiogram (ECG) captures the heart's electrical signals, offering valuable information for diagnosing cardiac conditions. However, the

tutorialsarxiv-cs-ai

Localizing Task Recognition and Task Learning in In-Context Learning via Attention Head Analysis

arXiv:2509.24164v2 Announce Type: replace Abstract: We investigate the mechanistic underpinnings of in-context learning (ICL) in large language models by reconciling two dominant perspectives: the com

researcharxiv-cs-cl

LuMon: A Comprehensive Benchmark and Development Suite with Novel Datasets for Lunar Monocular Depth Estimation

arXiv:2604.09352v1 Announce Type: new Abstract: Monocular Depth Estimation (MDE) is crucial for autonomous lunar rover navigation using electro-optical cameras. However, deploying terrestrial MDE netw

MATCHA: Efficient Deployment of Deep Neural Networks on Multi-Accelerator Heterogeneous Edge SoCs

arXiv:2604.09124v1 Announce Type: cross Abstract: Deploying DNNs on System-on-Chips (SoC) with multiple heterogeneous acceleration engines is challenging, and the majority of deployment frameworks can

MedConceal: A Benchmark for Clinical Hidden-Concern Reasoning Under Partial Observability

model-releasesarxiv-cs-cl

arXiv:2604.08788v1 Announce Type: new Abstract: Patient-clinician communication is an asymmetric-information problem: patients often do not disclose fears, misconceptions, or practical barriers unless

MONETA: Multimodal Industry Classification through Geographic Information with Multi Agent Systems

arXiv:2604.07956v2 Announce Type: replace Abstract: Industry classification schemes are integral parts of public and corporate databases as they classify businesses based on economic activity. Due to

MuTSE: A Human-in-the-Loop Multi-use Text Simplification Evaluator

arXiv:2604.08947v1 Announce Type: cross Abstract: As Large Language Models (LLMs) become increasingly prevalent in text simplification, systematically evaluating their outputs across diverse prompting

PDE-regularized Dynamics-informed Diffusion with Uncertainty-aware Filtering for Long-Horizon Dynamics

arXiv:2604.09058v1 Announce Type: cross Abstract: Long-horizon spatiotemporal prediction remains a challenging problem due to cumulative errors, noise amplification, and the lack of physical consisten

PerMix-RLVR: Preserving Persona Expressivity under Verifiable-Reward Alignment

arXiv:2604.08986v1 Announce Type: cross Abstract: Persona prompting has been widely adopted to steer large language models (LLMs) behavior and improve their instruction performance by assigning specif

Local Ai

Physics-Informed Reinforcement Learning of Spatial Density Velocity Potentials for Map-Free Racing

arXiv:2604.09499v1 Announce Type: new Abstract: Autonomous racing without prebuilt maps is a grand challenge for embedded robotics that requires kinodynamic planning from instantaneous sensor data at

local-aiarxiv-cs-ro

R2G: A Multi-View Circuit Graph Benchmark Suite from RTL to GDSII

arXiv:2604.08810v1 Announce Type: new Abstract: Graph neural networks (GNNs) are increasingly applied to physical design tasks such as congestion prediction and wirelength estimation, yet progress is

Retrieval Augmented Classification for Confidential Documents

arXiv:2604.08628v1 Announce Type: cross Abstract: Unauthorized disclosure of confidential documents demands robust, low-leakage classification. In real work environments, there is a lot of inflow and

StaRPO: Stability-Augmented Reinforcement Policy Optimization

arXiv:2604.08905v1 Announce Type: new Abstract: Reinforcement learning (RL) is effective in enhancing the accuracy of large language models in complex reasoning tasks. Existing RL policy optimization

Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insight

arXiv:2509.24169v2 Announce Type: replace Abstract: Large Language Models (LLMs) can perform new tasks from in-context demonstrations, a phenomenon known as in-context learning (ICL). Recent work sugg

researcharxiv-cs-cl

Text-Conditioned Multi-Expert Regression Framework for Fully Automated Multi-Abutment Design

arXiv:2604.09047v1 Announce Type: new Abstract: Dental implant abutments serve as the geometric and biomechanical interface between the implant fixture and the prosthetic crown, yet their design relie

TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation

arXiv:2604.09439v1 Announce Type: cross Abstract: In this paper, we propose a sequential recommendation model that integrates Time-aware personalization, Multi-interest personalization, and Explanatio

Towards Linguistically-informed Representations for English as a Second or Foreign Language: Review, Construction and Application

arXiv:2604.09008v1 Announce Type: cross Abstract: The widespread use of English as a Second or Foreign Language (ESFL) has sparked a paradigm shift: ESFL is not seen merely as a deviation from standar

Training event-based neural networks with exact gradients via Differentiable ODE Solving in JAX

arXiv:2603.08146v3 Announce Type: replace Abstract: Existing frameworks for gradient-based training of spiking neural networks face a trade-off: discrete-time methods using surrogate gradients support

safetyarxiv-cs-lg

A Data-Informed Variational Clustering Framework for Noisy High-Dimensional Data

arXiv:2604.06864v1 Announce Type: cross Abstract: Clustering in high-dimensional settings with severe feature noise remains challenging, especially when only a small subset of dimensions is informativ

Tutorials

A Severity-Based Curriculum Learning Strategy for Arabic Medical Text Generation

arXiv:2604.06365v1 Announce Type: cross Abstract: Arabic medical text generation is increasingly needed to help users interpret symptoms and access general health guidance in their native language. Ne

tutorialsarxiv-cs-ai

A Spatial-Spectral-Frequency Interactive Network for Multimodal Remote Sensing Classification

arXiv:2510.04628v2 Announce Type: replace Abstract: Deep learning-based methods have achieved significant success in remote sensing Earth observation data analysis. Numerous feature fusion techniques

Adaptive Depth-converted-Scale Convolution for Self-supervised Monocular Depth Estimation

arXiv:2604.07665v1 Announce Type: new Abstract: Self-supervised monocular depth estimation (MDE) has received increasing interests in the last few years. The objects in the scene, including the object

Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programs

arXiv:2604.06699v1 Announce Type: cross Abstract: Automated prompt optimization is crucial for eliciting reliable reasoning from large language models (LLMs), yet most API-only prompt optimizers itera

Tutorials

Analyzing Multimodal Interaction Strategies for LLM-Assisted Manipulation of 3D Scenes

arXiv:2410.22177v2 Announce Type: replace-cross Abstract: As more applications of large language models (LLMs) for 3D content for immersive environments emerge, it is crucial to study user behaviour t

tutorialsarxiv-cs-ai

ASBench: Image Anomalies Synthesis Benchmark for Anomaly Detection

arXiv:2510.07927v2 Announce Type: replace Abstract: Anomaly detection plays a pivotal role in manufacturing quality control, yet its application is constrained by limited abnormal samples and high man

AV-SQL: Decomposing Complex Text-to-SQL Queries with Agentic Views

arXiv:2604.07041v1 Announce Type: cross Abstract: Text-to-SQL is the task of translating natural language queries into executable SQL for a given database, enabling non-expert users to access structur

Bias Redistribution in Visual Machine Unlearning: Does Forgetting One Group Harm Another?

arXiv:2604.08111v1 Announce Type: cross Abstract: Machine unlearning enables models to selectively forget training data, driven by privacy regulations such as GDPR and CCPA. However, its fairness impl

safetyarxiv-cs-cv

Busemann energy-based attention for emotion analysis in Poincare discs

arXiv:2604.06752v1 Announce Type: new Abstract: We present EmBolic - a novel fully hyperbolic deep learning architecture for fine-grained emotion analysis from textual messages. The underlying idea is

Agents

Can VLMs Unlock Semantic Anomaly Detection? A Framework for Structured Reasoning

arXiv:2510.18034v2 Announce Type: replace-cross Abstract: Autonomous driving systems remain critically vulnerable to the long-tail of rare, out-of-distribution semantic anomalies. While VLMs have emer

agentsarxiv-cs-ai

Local Ai

CHiQPM: Calibrated Hierarchical Interpretable Image Classification

arXiv:2511.20779v2 Announce Type: replace Abstract: Globally interpretable models are a promising approach for trustworthy AI in safety-critical domains. Alongside global explanations, detailed local

local-aiarxiv-cs-lg

ChopGrad: Pixel-Wise Losses for Latent Video Diffusion via Truncated Backpropagation

arXiv:2603.17812v2 Announce Type: replace-cross Abstract: Recent video diffusion models achieve high-quality generation through recurrent frame processing where each frame generation depends on previo

ConvoLearn: A Dataset for Fine-Tuning Dialogic AI Tutors

arXiv:2601.08950v3 Announce Type: replace Abstract: Despite their growing adoption in education, LLMs remain misaligned with the core principle of effective tutoring: the dialogic construction of know

Decompose, Look, and Reason: Reinforced Latent Reasoning for VLMs

arXiv:2604.07518v1 Announce Type: new Abstract: Vision-Language Models often struggle with complex visual reasoning due to the visual information loss in textual CoT. Existing methods either add the c

safetyarxiv-cs-cl

Domain-Contextualized Inference: A Computable Graph Architecture for Explicit-Domain Reasoning

arXiv:2604.04344v2 Announce Type: replace Abstract: We establish a computation-substrate-agnostic inference architecture in which domain is an explicit first-class computational parameter. This produc

Drifting Fields are not Conservative

arXiv:2604.06333v1 Announce Type: new Abstract: Drifting models generate high-quality samples in a single forward pass by transporting generated samples toward the data distribution using a vector val