A Coding Implementation to Build and Train Advanced Architectures with Residual Connections, Self-Attention, and Adaptive Optimization Using JAX, Flax, and OptaxMarkTechPost In this tutorial, we explore how to build and train an advanced neural network using JAX, Flax, and Optax in an efficient and modular way. We begin by designing a deep architecture that integrates residual connections and self-attention mechanisms for expressive feature learning. As we progress, we implement sophisticated optimization strategies with learning rate scheduling,
The post A Coding Implementation to Build and Train Advanced Architectures with Residual Connections, Self-Attention, and Adaptive Optimization Using JAX, Flax, and Optax appeared first on MarkTechPost.
In this tutorial, we explore how to build and train an advanced neural network using JAX, Flax, and Optax in an efficient and modular way. We begin by designing a deep architecture that integrates residual connections and self-attention mechanisms for expressive feature learning. As we progress, we implement sophisticated optimization strategies with learning rate scheduling,
The post A Coding Implementation to Build and Train Advanced Architectures with Residual Connections, Self-Attention, and Adaptive Optimization Using JAX, Flax, and Optax appeared first on MarkTechPost. Read More
Rethinking Metrics and Diffusion Architecture for 3D Point Cloud Generationcs.AI updates on arXiv.org arXiv:2511.05308v2 Announce Type: cross
Abstract: As 3D point clouds become a cornerstone of modern technology, the need for sophisticated generative models and reliable evaluation metrics has grown exponentially. In this work, we first expose that some commonly used metrics for evaluating generated point clouds, particularly those based on Chamfer Distance (CD), lack robustness against defects and fail to capture geometric fidelity and local shape consistency when used as quality indicators. We further show that introducing samples alignment prior to distance calculation and replacing CD with Density-Aware Chamfer Distance (DCD) are simple yet essential steps to ensure the consistency and robustness of point cloud generative model evaluation metrics. While existing metrics primarily focus on directly comparing 3D Euclidean coordinates, we present a novel metric, named Surface Normal Concordance (SNC), which approximates surface similarity by comparing estimated point normals. This new metric, when combined with traditional ones, provides a more comprehensive evaluation of the quality of generated samples. Finally, leveraging recent advancements in transformer-based models for point cloud analysis, such as serialized patch attention , we propose a new architecture for generating high-fidelity 3D structures, the Diffusion Point Transformer. We perform extensive experiments and comparisons on the ShapeNet dataset, showing that our model outperforms previous solutions, particularly in terms of quality of generated point clouds, achieving new state-of-the-art. Code available at https://github.com/matteo-bastico/DiffusionPointTransformer.
arXiv:2511.05308v2 Announce Type: cross
Abstract: As 3D point clouds become a cornerstone of modern technology, the need for sophisticated generative models and reliable evaluation metrics has grown exponentially. In this work, we first expose that some commonly used metrics for evaluating generated point clouds, particularly those based on Chamfer Distance (CD), lack robustness against defects and fail to capture geometric fidelity and local shape consistency when used as quality indicators. We further show that introducing samples alignment prior to distance calculation and replacing CD with Density-Aware Chamfer Distance (DCD) are simple yet essential steps to ensure the consistency and robustness of point cloud generative model evaluation metrics. While existing metrics primarily focus on directly comparing 3D Euclidean coordinates, we present a novel metric, named Surface Normal Concordance (SNC), which approximates surface similarity by comparing estimated point normals. This new metric, when combined with traditional ones, provides a more comprehensive evaluation of the quality of generated samples. Finally, leveraging recent advancements in transformer-based models for point cloud analysis, such as serialized patch attention , we propose a new architecture for generating high-fidelity 3D structures, the Diffusion Point Transformer. We perform extensive experiments and comparisons on the ShapeNet dataset, showing that our model outperforms previous solutions, particularly in terms of quality of generated point clouds, achieving new state-of-the-art. Code available at https://github.com/matteo-bastico/DiffusionPointTransformer. Read More
Data Culture Is the Symptom, Not the SolutionTowards Data Science The hidden reason your data investments fail
The post Data Culture Is the Symptom, Not the Solution appeared first on Towards Data Science.
The hidden reason your data investments fail
The post Data Culture Is the Symptom, Not the Solution appeared first on Towards Data Science. Read More
Moonshot AI Releases Kosong: The LLM Abstraction Layer that Powers Kimi CLIMarkTechPost Modern agentic applications rarely talk to a single model or a single tool, so how do you keep that stack maintainable when providers, models and tools keep changing every few weeks. Moonshot AI’s Kosong targets this problem as an LLM abstraction layer for agent applications. Kosong unifies message structures, asynchronous tool orchestration and pluggable chat
The post Moonshot AI Releases Kosong: The LLM Abstraction Layer that Powers Kimi CLI appeared first on MarkTechPost.
Modern agentic applications rarely talk to a single model or a single tool, so how do you keep that stack maintainable when providers, models and tools keep changing every few weeks. Moonshot AI’s Kosong targets this problem as an LLM abstraction layer for agent applications. Kosong unifies message structures, asynchronous tool orchestration and pluggable chat
The post Moonshot AI Releases Kosong: The LLM Abstraction Layer that Powers Kimi CLI appeared first on MarkTechPost. Read More
Gelato-30B-A3B: A State-of-the-Art Grounding Model for GUI Computer-Use Tasks, Surpassing Computer Grounding Models like GTA1-32B MarkTechPost How do we teach AI agents to reliably find and click the exact on screen element we mean when we give them a simple instruction? A team of researchers from ML Foundations has introduced Gelato-30B-A3B, a state of the art grounding model for graphical user interfaces that is designed to plug into computer use agents
The post Gelato-30B-A3B: A State-of-the-Art Grounding Model for GUI Computer-Use Tasks, Surpassing Computer Grounding Models like GTA1-32B appeared first on MarkTechPost.
How do we teach AI agents to reliably find and click the exact on screen element we mean when we give them a simple instruction? A team of researchers from ML Foundations has introduced Gelato-30B-A3B, a state of the art grounding model for graphical user interfaces that is designed to plug into computer use agents
The post Gelato-30B-A3B: A State-of-the-Art Grounding Model for GUI Computer-Use Tasks, Surpassing Computer Grounding Models like GTA1-32B appeared first on MarkTechPost. Read More
Fine-tune VLMs for multipage document-to-JSON with SageMaker AI and SWIFTArtificial Intelligence In this post, we demonstrate that fine-tuning VLMs provides a powerful and flexible approach to automate and significantly enhance document understanding capabilities. We also demonstrate that using focused fine-tuning allows smaller, multi-modal models to compete effectively with much larger counterparts (98% accuracy with Qwen2.5 VL 3B).
In this post, we demonstrate that fine-tuning VLMs provides a powerful and flexible approach to automate and significantly enhance document understanding capabilities. We also demonstrate that using focused fine-tuning allows smaller, multi-modal models to compete effectively with much larger counterparts (98% accuracy with Qwen2.5 VL 3B). Read More
Make Python Up to 150× Faster with CTowards Data Science A practical guide to offloading performance-critical code to C without abandoning Python.
The post Make Python Up to 150× Faster with C appeared first on Towards Data Science.
A practical guide to offloading performance-critical code to C without abandoning Python.
The post Make Python Up to 150× Faster with C appeared first on Towards Data Science. Read More
Cognitive Edge Computing: A Comprehensive Survey on Optimizing Large Models and AI Agents for Pervasive Deploymentcs.AI updates on arXiv.org arXiv:2501.03265v2 Announce Type: replace-cross
Abstract: This article surveys Cognitive Edge Computing as a practical and methodical pathway for deploying reasoning-capable Large Language Models (LLMs) and autonomous AI agents on resource-constrained devices at the network edge. We present a unified, cognition-preserving framework spanning: (1) model optimization (quantization, sparsity, low-rank adaptation, distillation) aimed at retaining multi-step reasoning under tight memory/compute budgets; (2) system architecture (on-device inference, elastic offloading, cloud-edge collaboration) that trades off latency, energy, privacy, and capacity; and (3) adaptive intelligence (context compression, dynamic routing, federated personalization) that tailors computation to task difficulty and device constraints. We synthesize advances in efficient Transformer design, multimodal integration, hardware-aware compilation, privacy-preserving learning, and agentic tool use, and map them to edge-specific operating envelopes. We further outline a standardized evaluation protocol covering latency, throughput, energy per token, accuracy, robustness, privacy, and sustainability, with explicit measurement assumptions to enhance comparability. Remaining challenges include modality-aware reasoning benchmarks, transparent and reproducible energy reporting, edge-oriented safety/alignment evaluation, and multi-agent testbeds. We conclude with practitioner guidelines for cross-layer co-design of algorithms, runtime, and hardware to deliver reliable, efficient, and privacy-preserving cognitive capabilities on edge devices.
arXiv:2501.03265v2 Announce Type: replace-cross
Abstract: This article surveys Cognitive Edge Computing as a practical and methodical pathway for deploying reasoning-capable Large Language Models (LLMs) and autonomous AI agents on resource-constrained devices at the network edge. We present a unified, cognition-preserving framework spanning: (1) model optimization (quantization, sparsity, low-rank adaptation, distillation) aimed at retaining multi-step reasoning under tight memory/compute budgets; (2) system architecture (on-device inference, elastic offloading, cloud-edge collaboration) that trades off latency, energy, privacy, and capacity; and (3) adaptive intelligence (context compression, dynamic routing, federated personalization) that tailors computation to task difficulty and device constraints. We synthesize advances in efficient Transformer design, multimodal integration, hardware-aware compilation, privacy-preserving learning, and agentic tool use, and map them to edge-specific operating envelopes. We further outline a standardized evaluation protocol covering latency, throughput, energy per token, accuracy, robustness, privacy, and sustainability, with explicit measurement assumptions to enhance comparability. Remaining challenges include modality-aware reasoning benchmarks, transparent and reproducible energy reporting, edge-oriented safety/alignment evaluation, and multi-agent testbeds. We conclude with practitioner guidelines for cross-layer co-design of algorithms, runtime, and hardware to deliver reliable, efficient, and privacy-preserving cognitive capabilities on edge devices. Read More
To Trust or Not to Trust: On Calibration in ML-based Resource Allocation for Wireless Networkscs.AI updates on arXiv.org arXiv:2507.17494v2 Announce Type: replace-cross
Abstract: In next-generation communications and networks, machine learning (ML) models are expected to deliver not only accurate predictions but also well-calibrated confidence scores that reflect the true likelihood of correct decisions. This paper studies the calibration performance of an ML-based outage predictor within a single-user, multi-resource allocation framework. We first establish key theoretical properties of this system’s outage probability (OP) under perfect calibration. Importantly, we show that as the number of resources grows, the OP of a perfectly calibrated predictor approaches the expected output conditioned on it being below the classification threshold. In contrast, when only one resource is available, the system’s OP equals the model’s overall expected output. We then derive the OP conditions for a perfectly calibrated predictor. These findings guide the choice of the classification threshold to achieve a desired OP, helping system designers meet specific reliability requirements. We also demonstrate that post-processing calibration cannot improve the system’s minimum achievable OP, as it does not introduce new information about future channel states. Additionally, we show that well-calibrated models are part of a broader class of predictors that necessarily improve OP. In particular, we establish a monotonicity condition that the accuracy-confidence function must satisfy for such improvement to occur. To demonstrate these theoretical properties, we conduct a rigorous simulation-based analysis using post-processing calibration techniques: Platt scaling and isotonic regression. As part of this framework, the predictor is trained using an outage loss function specifically designed for this system. Furthermore, this analysis is performed on Rayleigh fading channels with temporal correlation captured by Clarke’s 2D model, which accounts for receiver mobility.
arXiv:2507.17494v2 Announce Type: replace-cross
Abstract: In next-generation communications and networks, machine learning (ML) models are expected to deliver not only accurate predictions but also well-calibrated confidence scores that reflect the true likelihood of correct decisions. This paper studies the calibration performance of an ML-based outage predictor within a single-user, multi-resource allocation framework. We first establish key theoretical properties of this system’s outage probability (OP) under perfect calibration. Importantly, we show that as the number of resources grows, the OP of a perfectly calibrated predictor approaches the expected output conditioned on it being below the classification threshold. In contrast, when only one resource is available, the system’s OP equals the model’s overall expected output. We then derive the OP conditions for a perfectly calibrated predictor. These findings guide the choice of the classification threshold to achieve a desired OP, helping system designers meet specific reliability requirements. We also demonstrate that post-processing calibration cannot improve the system’s minimum achievable OP, as it does not introduce new information about future channel states. Additionally, we show that well-calibrated models are part of a broader class of predictors that necessarily improve OP. In particular, we establish a monotonicity condition that the accuracy-confidence function must satisfy for such improvement to occur. To demonstrate these theoretical properties, we conduct a rigorous simulation-based analysis using post-processing calibration techniques: Platt scaling and isotonic regression. As part of this framework, the predictor is trained using an outage loss function specifically designed for this system. Furthermore, this analysis is performed on Rayleigh fading channels with temporal correlation captured by Clarke’s 2D model, which accounts for receiver mobility. Read More
Real-Time Reasoning Agents in Evolving Environmentscs.AI updates on arXiv.org arXiv:2511.04898v1 Announce Type: new
Abstract: Agents in the real world must make not only logical but also timely judgments. This requires continuous awareness of the dynamic environment: hazards emerge, opportunities arise, and other agents act, while the agent’s reasoning is still unfolding. Despite advances in language model reasoning, existing approaches fail to account for this dynamic nature. We introduce real-time reasoning as a new problem formulation for agents in evolving environments and build Real-Time Reasoning Gym to demonstrate it. We study two paradigms for deploying language models in agents: (1) reactive agents, which employ language models with bounded reasoning computation for rapid responses, and (2) planning agents, which allow extended reasoning computation for complex problems. Our experiments show that even state-of-the-art models struggle with making logical and timely judgments in either paradigm. To address this limitation, we propose AgileThinker, which simultaneously engages both reasoning paradigms. AgileThinker consistently outperforms agents engaging only one reasoning paradigm as the task difficulty and time pressure rise, effectively balancing reasoning depth and response latency. Our work establishes real-time reasoning as a critical testbed for developing practical agents and provides a foundation for research in temporally constrained AI systems, highlighting a path toward real-time capable agents.
arXiv:2511.04898v1 Announce Type: new
Abstract: Agents in the real world must make not only logical but also timely judgments. This requires continuous awareness of the dynamic environment: hazards emerge, opportunities arise, and other agents act, while the agent’s reasoning is still unfolding. Despite advances in language model reasoning, existing approaches fail to account for this dynamic nature. We introduce real-time reasoning as a new problem formulation for agents in evolving environments and build Real-Time Reasoning Gym to demonstrate it. We study two paradigms for deploying language models in agents: (1) reactive agents, which employ language models with bounded reasoning computation for rapid responses, and (2) planning agents, which allow extended reasoning computation for complex problems. Our experiments show that even state-of-the-art models struggle with making logical and timely judgments in either paradigm. To address this limitation, we propose AgileThinker, which simultaneously engages both reasoning paradigms. AgileThinker consistently outperforms agents engaging only one reasoning paradigm as the task difficulty and time pressure rise, effectively balancing reasoning depth and response latency. Our work establishes real-time reasoning as a critical testbed for developing practical agents and provides a foundation for research in temporally constrained AI systems, highlighting a path toward real-time capable agents. Read More
