TJS Articles & News

AI memory hunger forces Micron’s consumer exodus: A turning point in semiconductor economicsAI News In the basement of a Boise, Idaho, dental office in 1978, four engineers founded what would become one of America’s semiconductor giants. Ward Parkinson, Joe Parkinson, Dennis Wilson, and Doug Pitman started Micron Technology as a modest design consultancy, backed by local investors including potato magnate J.R. Simplot. By 1983, they had achieved a technological
The post AI memory hunger forces Micron’s consumer exodus: A turning point in semiconductor economics appeared first on AI News.

 In the basement of a Boise, Idaho, dental office in 1978, four engineers founded what would become one of America’s semiconductor giants. Ward Parkinson, Joe Parkinson, Dennis Wilson, and Doug Pitman started Micron Technology as a modest design consultancy, backed by local investors including potato magnate J.R. Simplot. By 1983, they had achieved a technological
The post AI memory hunger forces Micron’s consumer exodus: A turning point in semiconductor economics appeared first on AI News. Read More  

GAPO: Robust Advantage Estimation for Real-World Code LLMscs.AI updates on arXiv.org arXiv:2510.21830v3 Announce Type: replace-cross
Abstract: Reinforcement learning (RL) is widely used for post-training large language models (LLMs) in code editing, where group-relative methods like GRPO are popular for their critic-free, normalized advantage estimation. However, in real-world code-editing scenarios, reward distributions are often skewed with unpredictable outliers, leading to distorted advantage computation and increased noise. To address this issue, we propose Group Adaptive Policy Optimization (GAPO), which adaptively finds an outlier-free highest-density interval (HDI) per prompt and then uses the median of that interval as an adaptive Q to replace the group mean in advantage calculation. This adaptive Q robustly handles skewed distributions while remaining plug-and-play and efficient. We validate GAPO on nine instruction-tuned LLMs (3B-14B) using a large internal dataset of 51,844 real-world, history-aware code-editing tasks across 10 languages, demonstrating consistent improvements in exact match accuracy over GRPO and its variant DAPO. Code is publicly available.

 arXiv:2510.21830v3 Announce Type: replace-cross
Abstract: Reinforcement learning (RL) is widely used for post-training large language models (LLMs) in code editing, where group-relative methods like GRPO are popular for their critic-free, normalized advantage estimation. However, in real-world code-editing scenarios, reward distributions are often skewed with unpredictable outliers, leading to distorted advantage computation and increased noise. To address this issue, we propose Group Adaptive Policy Optimization (GAPO), which adaptively finds an outlier-free highest-density interval (HDI) per prompt and then uses the median of that interval as an adaptive Q to replace the group mean in advantage calculation. This adaptive Q robustly handles skewed distributions while remaining plug-and-play and efficient. We validate GAPO on nine instruction-tuned LLMs (3B-14B) using a large internal dataset of 51,844 real-world, history-aware code-editing tasks across 10 languages, demonstrating consistent improvements in exact match accuracy over GRPO and its variant DAPO. Code is publicly available. Read More  

Just-in-time and distributed task representations in language modelscs.AI updates on arXiv.org arXiv:2509.04466v3 Announce Type: replace-cross
Abstract: Many of language models’ impressive capabilities originate from their in-context learning: based on instructions or examples, they can infer and perform new tasks without weight updates. In this work, we investigate when representations for new tasks are formed in language models, and how these representations change over the course of context. We study two different task representations: those that are ”transferrable” — vector representations that can transfer task contexts to another model instance, even without the full prompt — and simpler representations of high-level task categories. We show that transferrable task representations evolve in non-monotonic and sporadic ways, while task identity representations persist throughout the context. Specifically, transferrable task representations exhibit a two-fold locality. They successfully condense evidence when more examples are provided in the context. But this evidence accrual process exhibits strong temporal locality along the sequence dimension, coming online only at certain tokens — despite task identity being reliably decodable throughout the context. In some cases, transferrable task representations also show semantic locality, capturing a small task ”scope” such as an independent subtask. Language models thus represent new tasks on the fly through both an inert, sustained sensitivity to the task and an active, just-in-time representation to support inference.

 arXiv:2509.04466v3 Announce Type: replace-cross
Abstract: Many of language models’ impressive capabilities originate from their in-context learning: based on instructions or examples, they can infer and perform new tasks without weight updates. In this work, we investigate when representations for new tasks are formed in language models, and how these representations change over the course of context. We study two different task representations: those that are ”transferrable” — vector representations that can transfer task contexts to another model instance, even without the full prompt — and simpler representations of high-level task categories. We show that transferrable task representations evolve in non-monotonic and sporadic ways, while task identity representations persist throughout the context. Specifically, transferrable task representations exhibit a two-fold locality. They successfully condense evidence when more examples are provided in the context. But this evidence accrual process exhibits strong temporal locality along the sequence dimension, coming online only at certain tokens — despite task identity being reliably decodable throughout the context. In some cases, transferrable task representations also show semantic locality, capturing a small task ”scope” such as an independent subtask. Language models thus represent new tasks on the fly through both an inert, sustained sensitivity to the task and an active, just-in-time representation to support inference. Read More  

Knowledge Adaptation as Posterior Correctioncs.AI updates on arXiv.org arXiv:2506.14262v2 Announce Type: replace-cross
Abstract: Adaptation is the holy grail of intelligence, but even the best AI models lack the adaptability of toddlers. In spite of great progress, little is known about the mechanisms by which machines can learn to adapt as fast as humans and animals. Here, we cast adaptation as `correction’ of old posteriors and show that a wide-variety of existing adaptation methods follow this very principle, including those used for continual learning, federated learning, unlearning, and model merging. In all these settings, more accurate posteriors often lead to smaller corrections and can enable faster adaptation. Posterior correction is derived by using the dual representation of the Bayesian Learning Rule of Khan and Rue (2023), where the interference between the old representation and new information is quantified by using the natural-gradient mismatch. We present many examples demonstrating how machines can learn to adapt quickly by using posterior correction.

 arXiv:2506.14262v2 Announce Type: replace-cross
Abstract: Adaptation is the holy grail of intelligence, but even the best AI models lack the adaptability of toddlers. In spite of great progress, little is known about the mechanisms by which machines can learn to adapt as fast as humans and animals. Here, we cast adaptation as `correction’ of old posteriors and show that a wide-variety of existing adaptation methods follow this very principle, including those used for continual learning, federated learning, unlearning, and model merging. In all these settings, more accurate posteriors often lead to smaller corrections and can enable faster adaptation. Posterior correction is derived by using the dual representation of the Bayesian Learning Rule of Khan and Rue (2023), where the interference between the old representation and new information is quantified by using the natural-gradient mismatch. We present many examples demonstrating how machines can learn to adapt quickly by using posterior correction. Read More  

Multi-Agent Arena: Insights from London Great Agent Hack 2025Towards Data Science What mattered: robust agents, glass-box reasoning, and red-team resilience
The post Multi-Agent Arena: Insights from London Great Agent Hack 2025 appeared first on Towards Data Science.

 What mattered: robust agents, glass-box reasoning, and red-team resilience
The post Multi-Agent Arena: Insights from London Great Agent Hack 2025 appeared first on Towards Data Science. Read More  

10 GitHub Repositories to Master Vibe CodingKDnuggets Explore top GitHub repositories to help you master this new style of coding and ship full-stack products faster than ever.

 Explore top GitHub repositories to help you master this new style of coding and ship full-stack products faster than ever. Read More  

How to Code Your Own Website with AITowards Data Science Learn how to vibe-code your own website
The post How to Code Your Own Website with AI appeared first on Towards Data Science.

 Learn how to vibe-code your own website
The post How to Code Your Own Website with AI appeared first on Towards Data Science. Read More  

EY and NVIDIA to help companies test and deploy physical AIAI News AI is moving deeper into the physical world, and EY is laying out a more structured way for companies to work with robots, drones, and other smart devices. The organisation is introducing a physical AI platform built with NVIDIA tools, opening a new EY.ai Lab in Georgia, and adding new leadership to guide its work
The post EY and NVIDIA to help companies test and deploy physical AI appeared first on AI News.

 AI is moving deeper into the physical world, and EY is laying out a more structured way for companies to work with robots, drones, and other smart devices. The organisation is introducing a physical AI platform built with NVIDIA tools, opening a new EY.ai Lab in Georgia, and adding new leadership to guide its work
The post EY and NVIDIA to help companies test and deploy physical AI appeared first on AI News. Read More  

How to Turn Your LLM Prototype into a Production-Ready SystemTowards Data Science The most famous applications of LLMs are the ones that I like to call the “wow effect LLMs.” There are plenty of viral LinkedIn posts about them, and they all sound like this: “I built [x] that does [y] in [z] minutes using AI.” Where: If you notice carefully, the focus of the sentence is
The post How to Turn Your LLM Prototype into a Production-Ready System appeared first on Towards Data Science.

 The most famous applications of LLMs are the ones that I like to call the “wow effect LLMs.” There are plenty of viral LinkedIn posts about them, and they all sound like this: “I built [x] that does [y] in [z] minutes using AI.” Where: If you notice carefully, the focus of the sentence is
The post How to Turn Your LLM Prototype into a Production-Ready System appeared first on Towards Data Science. Read More  

HTB AI Range offers experiments in cyber-resilience trainingAI News The cybersecurity training provider Hack The Box (HTB) has launched the HTB AI Range, designed to let organisations test autonomous AI security agents under realistic conditions, albeit with oversight from human cybersecurity professionals. Its goal is to help users assess how well AI, and mixed human–AI teams might defend infrastructure. Vulnerabilities in AI models add
The post HTB AI Range offers experiments in cyber-resilience training appeared first on AI News.

 The cybersecurity training provider Hack The Box (HTB) has launched the HTB AI Range, designed to let organisations test autonomous AI security agents under realistic conditions, albeit with oversight from human cybersecurity professionals. Its goal is to help users assess how well AI, and mixed human–AI teams might defend infrastructure. Vulnerabilities in AI models add
The post HTB AI Range offers experiments in cyber-resilience training appeared first on AI News. Read More