NCSAM Noise-Compensated Sharpness-Aware Minimization for Noisy Label Learningcs.AI updates on arXiv.org arXiv:2601.19947v1 Announce Type: cross
Abstract: Learning from Noisy Labels (LNL) presents a fundamental challenge in deep learning, as real-world datasets often contain erroneous or corrupted annotations, textit{e.g.}, data crawled from Web. Current research focuses on sophisticated label correction mechanisms. In contrast, this paper adopts a novel perspective by establishing a theoretical analysis the relationship between flatness of the loss landscape and the presence of label noise. In this paper, we theoretically demonstrate that carefully simulated label noise synergistically enhances both the generalization performance and robustness of label noises. Consequently, we propose Noise-Compensated Sharpness-aware Minimization (NCSAM) to leverage the perturbation of Sharpness-Aware Minimization (SAM) to remedy the damage of label noises. Our analysis reveals that the testing accuracy exhibits a similar behavior that has been observed on the noise-clear dataset. Extensive experimental results on multiple benchmark datasets demonstrate the consistent superiority of the proposed method over existing state-of-the-art approaches on diverse tasks.
arXiv:2601.19947v1 Announce Type: cross
Abstract: Learning from Noisy Labels (LNL) presents a fundamental challenge in deep learning, as real-world datasets often contain erroneous or corrupted annotations, textit{e.g.}, data crawled from Web. Current research focuses on sophisticated label correction mechanisms. In contrast, this paper adopts a novel perspective by establishing a theoretical analysis the relationship between flatness of the loss landscape and the presence of label noise. In this paper, we theoretically demonstrate that carefully simulated label noise synergistically enhances both the generalization performance and robustness of label noises. Consequently, we propose Noise-Compensated Sharpness-aware Minimization (NCSAM) to leverage the perturbation of Sharpness-Aware Minimization (SAM) to remedy the damage of label noises. Our analysis reveals that the testing accuracy exhibits a similar behavior that has been observed on the noise-clear dataset. Extensive experimental results on multiple benchmark datasets demonstrate the consistent superiority of the proposed method over existing state-of-the-art approaches on diverse tasks. Read More
5 Fun APIs for Absolute BeginnersKDnuggets Five APIs that make experimenting with AI and web data easy, practical, and beginner-friendly.
Five APIs that make experimenting with AI and web data easy, practical, and beginner-friendly. Read More
AI use surges at Travelers as call centre roles reduceAI News Mid-January saw insurance company, Travelers, announce a new deal that empowers 10,000 engineers and data scientists with AI assistants. However, less than two weeks on, Travelers’ leadership explained that the company’s true competitive advantage lies in expertise, not AIs alone, believing this is what will drive longer-term profit growth. According to Travelers’ chief executive officer
The post AI use surges at Travelers as call centre roles reduce appeared first on AI News.
Mid-January saw insurance company, Travelers, announce a new deal that empowers 10,000 engineers and data scientists with AI assistants. However, less than two weeks on, Travelers’ leadership explained that the company’s true competitive advantage lies in expertise, not AIs alone, believing this is what will drive longer-term profit growth. According to Travelers’ chief executive officer
The post AI use surges at Travelers as call centre roles reduce appeared first on AI News. Read More
PepsiCo is using AI to rethink how factories are designed and updatedAI News For many large companies, the most useful form of AI right now has little to do with writing emails or answering questions. At PepsiCo, AI is being tested in places where mistakes are costly and changes are hard to undo — factory layouts, production lines, and physical operations. That shift is visible in how PepsiCo
The post PepsiCo is using AI to rethink how factories are designed and updated appeared first on AI News.
For many large companies, the most useful form of AI right now has little to do with writing emails or answering questions. At PepsiCo, AI is being tested in places where mistakes are costly and changes are hard to undo — factory layouts, production lines, and physical operations. That shift is visible in how PepsiCo
The post PepsiCo is using AI to rethink how factories are designed and updated appeared first on AI News. Read More
Microsoft Unveils Maia 200, An FP4 and FP8 Optimized AI Inference Accelerator for Azure DatacentersMarkTechPost Maia 200 is Microsoft’s new in house AI accelerator designed for inference in Azure datacenters. It targets the cost of token generation for large language models and other reasoning workloads by combining narrow precision compute, a dense on chip memory hierarchy and an Ethernet based scale up fabric. Why Microsoft built a dedicated inference chip?
The post Microsoft Unveils Maia 200, An FP4 and FP8 Optimized AI Inference Accelerator for Azure Datacenters appeared first on MarkTechPost.
Maia 200 is Microsoft’s new in house AI accelerator designed for inference in Azure datacenters. It targets the cost of token generation for large language models and other reasoning workloads by combining narrow precision compute, a dense on chip memory hierarchy and an Ethernet based scale up fabric. Why Microsoft built a dedicated inference chip?
The post Microsoft Unveils Maia 200, An FP4 and FP8 Optimized AI Inference Accelerator for Azure Datacenters appeared first on MarkTechPost. Read More
China’s hyperscalers bet billions on agentic AI as commerce becomes the new battlegroundAI News The artificial intelligence industry’s pivot toward agentic AI – systems capable of autonomously executing multi-step tasks – has dominated technology discussions in recent months. But while Western firms focus on foundational models and cross-platform interoperability, China’s technology giants are racing to dominate through commerce integration, a divergence that could reshape how enterprises deploy autonomous systems
The post China’s hyperscalers bet billions on agentic AI as commerce becomes the new battleground appeared first on AI News.
The artificial intelligence industry’s pivot toward agentic AI – systems capable of autonomously executing multi-step tasks – has dominated technology discussions in recent months. But while Western firms focus on foundational models and cross-platform interoperability, China’s technology giants are racing to dominate through commerce integration, a divergence that could reshape how enterprises deploy autonomous systems
The post China’s hyperscalers bet billions on agentic AI as commerce becomes the new battleground appeared first on AI News. Read More
DeepSeek AI Releases DeepSeek-OCR 2 with Causal Visual Flow Encoder for Layout Aware Document UnderstandingMarkTechPost DeepSeek AI released DeepSeek-OCR 2, an open source document OCR and understanding system that restructures its vision encoder to read pages in a causal order that is closer to how humans scan complex documents. The key component is DeepEncoder V2, a language model style transformer that converts a 2D page into a 1D sequence of
The post DeepSeek AI Releases DeepSeek-OCR 2 with Causal Visual Flow Encoder for Layout Aware Document Understanding appeared first on MarkTechPost.
DeepSeek AI released DeepSeek-OCR 2, an open source document OCR and understanding system that restructures its vision encoder to read pages in a causal order that is closer to how humans scan complex documents. The key component is DeepEncoder V2, a language model style transformer that converts a 2D page into a 1D sequence of
The post DeepSeek AI Releases DeepSeek-OCR 2 with Causal Visual Flow Encoder for Layout Aware Document Understanding appeared first on MarkTechPost. Read More
A Coding Deep Dive into Differentiable Computer Vision with Kornia Using Geometry Optimization, LoFTR Matching, and GPU AugmentationsMarkTechPost We implement an advanced, end-to-end Kornia tutorial and demonstrate how modern, differentiable computer vision can be built entirely in PyTorch. We start by constructing GPU-accelerated, synchronized augmentation pipelines for images, masks, and keypoints, then move into differentiable geometry by optimizing a homography directly through gradient descent. We also show how learned feature matching with LoFTR
The post A Coding Deep Dive into Differentiable Computer Vision with Kornia Using Geometry Optimization, LoFTR Matching, and GPU Augmentations appeared first on MarkTechPost.
We implement an advanced, end-to-end Kornia tutorial and demonstrate how modern, differentiable computer vision can be built entirely in PyTorch. We start by constructing GPU-accelerated, synchronized augmentation pipelines for images, masks, and keypoints, then move into differentiable geometry by optimizing a homography directly through gradient descent. We also show how learned feature matching with LoFTR
The post A Coding Deep Dive into Differentiable Computer Vision with Kornia Using Geometry Optimization, LoFTR Matching, and GPU Augmentations appeared first on MarkTechPost. Read More
Agent Benchmarks Fail Public Sector Requirementscs.AI updates on arXiv.org arXiv:2601.20617v1 Announce Type: cross
Abstract: Deploying Large Language Model-based agents (LLM agents) in the public sector requires assuring that they meet the stringent legal, procedural, and structural requirements of public-sector institutions. Practitioners and researchers often turn to benchmarks for such assessments. However, it remains unclear what criteria benchmarks must meet to ensure they adequately reflect public-sector requirements, or how many existing benchmarks do so. In this paper, we first define such criteria based on a first-principles survey of public administration literature: benchmarks must be emph{process-based}, emph{realistic}, emph{public-sector-specific} and report emph{metrics} that reflect the unique requirements of the public sector. We analyse more than 1,300 benchmark papers for these criteria using an expert-validated LLM-assisted pipeline. Our results show that no single benchmark meets all of the criteria. Our findings provide a call to action for both researchers to develop public sector-relevant benchmarks and for public-sector officials to apply these criteria when evaluating their own agentic use cases.
arXiv:2601.20617v1 Announce Type: cross
Abstract: Deploying Large Language Model-based agents (LLM agents) in the public sector requires assuring that they meet the stringent legal, procedural, and structural requirements of public-sector institutions. Practitioners and researchers often turn to benchmarks for such assessments. However, it remains unclear what criteria benchmarks must meet to ensure they adequately reflect public-sector requirements, or how many existing benchmarks do so. In this paper, we first define such criteria based on a first-principles survey of public administration literature: benchmarks must be emph{process-based}, emph{realistic}, emph{public-sector-specific} and report emph{metrics} that reflect the unique requirements of the public sector. We analyse more than 1,300 benchmark papers for these criteria using an expert-validated LLM-assisted pipeline. Our results show that no single benchmark meets all of the criteria. Our findings provide a call to action for both researchers to develop public sector-relevant benchmarks and for public-sector officials to apply these criteria when evaluating their own agentic use cases. Read More
Unsupervised Ensemble Learning Through Deep Energy-based Modelscs.AI updates on arXiv.org arXiv:2601.20556v1 Announce Type: cross
Abstract: Unsupervised ensemble learning emerged to address the challenge of combining multiple learners’ predictions without access to ground truth labels or additional data. This paradigm is crucial in scenarios where evaluating individual classifier performance or understanding their strengths is challenging due to limited information. We propose a novel deep energy-based method for constructing an accurate meta-learner using only the predictions of individual learners, potentially capable of capturing complex dependence structures between them. Our approach requires no labeled data, learner features, or problem-specific information, and has theoretical guarantees for when learners are conditionally independent. We demonstrate superior performance across diverse ensemble scenarios, including challenging mixture of experts settings. Our experiments span standard ensemble datasets and curated datasets designed to test how the model fuses expertise from multiple sources. These results highlight the potential of unsupervised ensemble learning to harness collective intelligence, especially in data-scarce or privacy-sensitive environments.
arXiv:2601.20556v1 Announce Type: cross
Abstract: Unsupervised ensemble learning emerged to address the challenge of combining multiple learners’ predictions without access to ground truth labels or additional data. This paradigm is crucial in scenarios where evaluating individual classifier performance or understanding their strengths is challenging due to limited information. We propose a novel deep energy-based method for constructing an accurate meta-learner using only the predictions of individual learners, potentially capable of capturing complex dependence structures between them. Our approach requires no labeled data, learner features, or problem-specific information, and has theoretical guarantees for when learners are conditionally independent. We demonstrate superior performance across diverse ensemble scenarios, including challenging mixture of experts settings. Our experiments span standard ensemble datasets and curated datasets designed to test how the model fuses expertise from multiple sources. These results highlight the potential of unsupervised ensemble learning to harness collective intelligence, especially in data-scarce or privacy-sensitive environments. Read More
