AI Infrastructure

GPUs, training clusters, MLOps, and deployment

This Week's Best | Monthly Best | Guide | Trending

Top This Week

Machine Learning

Trained a Qwen2.5-0.5B-Instruct bf16 model on Reddit post summarization task with GRPO written from scratch in PyTorch - updates! [P]

So, yesterday run was a success and I did get an avg rollout length of about 64 tokens as attached in the image! This was with quality_re...

Reddit - Machine Learning · 1 min ·
[2603.10652] Are Video Reasoning Models Ready to Go Outside?
Llms

[2603.10652] Are Video Reasoning Models Ready to Go Outside?

Abstract page for arXiv paper 2603.10652: Are Video Reasoning Models Ready to Go Outside?

arXiv - AI · 4 min ·
[2602.00181] CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning
Machine Learning

[2602.00181] CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning

Abstract page for arXiv paper 2602.00181: CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning

arXiv - AI · 4 min ·

All Content

[2602.15379] FlashMem: Supporting Modern DNN Workloads on Mobile with GPU Memory Hierarchy Optimizations
Machine Learning

[2602.15379] FlashMem: Supporting Modern DNN Workloads on Mobile with GPU Memory Hierarchy Optimizations

The paper presents FlashMem, a memory streaming framework designed to optimize the execution of large-scale deep neural networks (DNNs) o...

arXiv - Machine Learning · 4 min ·
[2602.15669] PERSONA: Dynamic and Compositional Inference-Time Personality Control via Activation Vector Algebra
Llms

[2602.15669] PERSONA: Dynamic and Compositional Inference-Time Personality Control via Activation Vector Algebra

The paper introduces PERSONA, a novel framework for dynamic personality control in Large Language Models (LLMs) using activation vector a...

arXiv - AI · 4 min ·
[2602.15326] SCENE OTA-FD: Self-Centering Noncoherent Estimator for Over-the-Air Federated Distillation
Ai Safety

[2602.15326] SCENE OTA-FD: Self-Centering Noncoherent Estimator for Over-the-Air Federated Distillation

The paper presents SCENE, a novel estimator for over-the-air federated distillation that enhances aggregation without requiring pilot sig...

arXiv - Machine Learning · 3 min ·
[2602.15277] Accelerating Large-Scale Dataset Distillation via Exploration-Exploitation Optimization
Machine Learning

[2602.15277] Accelerating Large-Scale Dataset Distillation via Exploration-Exploitation Optimization

This paper presents Exploration-Exploitation Distillation (E^2D), a method for efficient large-scale dataset distillation that balances a...

arXiv - Machine Learning · 4 min ·
[2602.15391] Improving LLM Reliability through Hybrid Abstention and Adaptive Detection
Llms

[2602.15391] Improving LLM Reliability through Hybrid Abstention and Adaptive Detection

The paper presents a novel adaptive abstention system for Large Language Models (LLMs) that balances safety and utility by dynamically ad...

arXiv - AI · 4 min ·
[2602.15161] Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning
Machine Learning

[2602.15161] Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning

This paper presents the Layer Smoothing Attack (LSA), a novel backdoor attack exploiting layer-specific vulnerabilities in federated lear...

arXiv - Machine Learning · 4 min ·
[2602.15136] Universal priors: solving empirical Bayes via Bayesian inference and pretraining
Machine Learning

[2602.15136] Universal priors: solving empirical Bayes via Bayesian inference and pretraining

The paper explores how a pretrained transformer can effectively solve empirical Bayes problems by leveraging universal priors, demonstrat...

arXiv - Machine Learning · 3 min ·
[2602.15248] Predicting Invoice Dilution in Supply Chain Finance with Leakage Free Two Stage XGBoost, KAN (Kolmogorov Arnold Networks), and Ensemble Models
Machine Learning

[2602.15248] Predicting Invoice Dilution in Supply Chain Finance with Leakage Free Two Stage XGBoost, KAN (Kolmogorov Arnold Networks), and Ensemble Models

This paper presents a machine learning framework to predict invoice dilution in supply chain finance, utilizing advanced models like XGBo...

arXiv - AI · 3 min ·
[2602.15084] TokaMind: A Multi-Modal Transformer Foundation Model for Tokamak Plasma Dynamics
Llms

[2602.15084] TokaMind: A Multi-Modal Transformer Foundation Model for Tokamak Plasma Dynamics

TokaMind is a new open-source multi-modal transformer model designed for tokamak plasma dynamics, demonstrating superior performance on f...

arXiv - Machine Learning · 4 min ·
[2602.15212] Secure and Energy-Efficient Wireless Agentic AI Networks
Ai Agents

[2602.15212] Secure and Energy-Efficient Wireless Agentic AI Networks

This paper presents a secure and energy-efficient wireless AI network that utilizes a supervisor AI agent to optimize reasoning tasks whi...

arXiv - AI · 4 min ·
[2602.15156] Panini: Continual Learning in Token Space via Structured Memory
Llms

[2602.15156] Panini: Continual Learning in Token Space via Structured Memory

The paper presents Panini, a continual learning framework for language models that enhances efficiency and accuracy by integrating experi...

arXiv - AI · 4 min ·
[2602.15820] Stabilizing Test-Time Adaptation of High-Dimensional Simulation Surrogates via D-Optimal Statistics
Machine Learning

[2602.15820] Stabilizing Test-Time Adaptation of High-Dimensional Simulation Surrogates via D-Optimal Statistics

This article presents a framework for Test-Time Adaptation (TTA) of high-dimensional simulation surrogates using D-optimal statistics, ad...

arXiv - Machine Learning · 3 min ·
[2602.15763] GLM-5: from Vibe Coding to Agentic Engineering
Llms

[2602.15763] GLM-5: from Vibe Coding to Agentic Engineering

GLM-5 introduces a next-generation foundation model that enhances coding capabilities through agentic engineering, reducing costs while i...

arXiv - Machine Learning · 5 min ·
[2602.15711] Random Wavelet Features for Graph Kernel Machines
Nlp

[2602.15711] Random Wavelet Features for Graph Kernel Machines

This paper introduces randomized spectral node embeddings for graph kernel machines, enhancing node similarity estimation while improving...

arXiv - AI · 3 min ·
[2602.15704] Controlled oscillation modeling using port-Hamiltonian neural networks
Machine Learning

[2602.15704] Controlled oscillation modeling using port-Hamiltonian neural networks

This paper presents a novel approach to modeling controlled oscillations using port-Hamiltonian neural networks, emphasizing a second-ord...

arXiv - Machine Learning · 4 min ·
[2602.15648] Guided Diffusion by Optimized Loss Functions on Relaxed Parameters for Inverse Material Design
Generative Ai

[2602.15648] Guided Diffusion by Optimized Loss Functions on Relaxed Parameters for Inverse Material Design

This paper presents a novel method for inverse material design using guided diffusion and optimized loss functions, addressing challenges...

arXiv - Machine Learning · 4 min ·
[2602.15617] DNN-Enabled Multi-User Beamforming for Throughput Maximization under Adjustable Fairness
Machine Learning

[2602.15617] DNN-Enabled Multi-User Beamforming for Throughput Maximization under Adjustable Fairness

This paper presents a DNN-based approach to optimize multi-user beamforming in wireless communications, balancing throughput and fairness...

arXiv - Machine Learning · 3 min ·
[2602.15571] Accelerated Predictive Coding Networks via Direct Kolen-Pollack Feedback Alignment
Machine Learning

[2602.15571] Accelerated Predictive Coding Networks via Direct Kolen-Pollack Feedback Alignment

The paper introduces Direct Kolen-Pollack Predictive Coding (DKP-PC), an innovative approach that enhances the efficiency of predictive c...

arXiv - Machine Learning · 3 min ·
[2602.15563] 1-Bit Wonder: Improving QAT Performance in the Low-Bit Regime through K-Means Quantization
Llms

[2602.15563] 1-Bit Wonder: Improving QAT Performance in the Low-Bit Regime through K-Means Quantization

The paper presents a study on quantization-aware training (QAT) for low-bit quantization, demonstrating that k-means based weight quantiz...

arXiv - Machine Learning · 3 min ·
[2602.15546] CEPAE: Conditional Entropy-Penalized Autoencoders for Time Series Counterfactuals
Machine Learning

[2602.15546] CEPAE: Conditional Entropy-Penalized Autoencoders for Time Series Counterfactuals

The paper introduces CEPAE, a novel approach utilizing Conditional Entropy-Penalized Autoencoders for effective counterfactual inference ...

arXiv - Machine Learning · 3 min ·
Previous Page 153 Next

Related Topics

Machine Learning Large Language Models Generative AI Natural Language Processing Computer Vision Robotics & Embodied AI

Stay updated with AI News

Get the latest news, tools, and insights delivered to your inbox.

Subscribe to Newsletter

Daily or weekly digest • Unsubscribe anytime