DeepSeek AI Researchers Introduce Engram: A Conditional Memory Axis For Sparse LLMs
Transformers use attention and Mixture-of-Experts to scale computation, but they still lack a native way to perform knowledge lookup. They re-compute the same local patterns again and again, which wastes depth and FLOPs. DeepSeek’s new Engram module targets exactly this gap by adding a conditional m...
Spectral Generative Flow Models: A Physics-Inspired Replacement for Vectorized Large Language Models
arXiv:2601.08893v1 Announce Type: new
Abstract: We introduce Spectral Generative Flow Models (SGFMs), a physics-inspired alternative to transformer-based large language models. Instead of representing text or video as sequences of discrete tokens processed by attention, SGFMs treat generation as th...
XGBoost Forecasting of NEPSE Index Log Returns with Walk Forward Validation
arXiv:2601.08896v1 Announce Type: new
Abstract: This study develops a robust machine learning framework for one-step-ahead forecasting of daily log-returns in the Nepal Stock Exchange (NEPSE) Index using the XGBoost regressor. A comprehensive feature set is engineered, including lagged log-returns ...
DriftGuard: A Hierarchical Framework for Concept Drift Detection and Remediation in Supply Chain Forecasting
arXiv:2601.08928v1 Announce Type: new
Abstract: Supply chain forecasting models degrade over time as real-world conditions change. Promotions shift, consumer preferences evolve, and supply disruptions alter demand patterns, causing what is known as concept drift. This silent degradation leads to st...
Breaking the Bottlenecks: Scalable Diffusion Models for 3D Molecular Generation
arXiv:2601.08963v1 Announce Type: new
Abstract: Diffusion models have emerged as a powerful class of generative models for molecular design, capable of capturing complex structural distributions and achieving high fidelity in 3D molecule generation. However, their widespread use remains constrained...
ConvoLearn: A Dataset of Constructivist Tutor-Student Dialogue
arXiv:2601.08950v1 Announce Type: new
Abstract: In educational applications, LLMs exhibit several fundamental pedagogical limitations, such as their tendency to reveal solutions rather than support dialogic learning. We introduce ConvoLearn (https://huggingface.co/datasets/masharma/convolearn ), a ...
ART: Action-based Reasoning Task Benchmarking for Medical AI Agents
arXiv:2601.08988v1 Announce Type: new
Abstract: Reliable clinical decision support requires medical AI agents capable of safe, multi-step reasoning over structured electronic health records (EHRs). While large language models (LLMs) show promise in healthcare, existing benchmarks inadequately asses...
The Hierarchy of Agentic Capabilities: Evaluating Frontier Models on Realistic RL Environments
arXiv:2601.09032v1 Announce Type: new
Abstract: The advancement of large language model (LLM) based agents has shifted AI evaluation from single-turn response assessment to multi-step task completion in interactive environments. We present an empirical study evaluating frontier AI models on 150 wor...
arXiv:2601.09072v1 Announce Type: new
Abstract: Developing safe, effective, and practically useful clinical prediction models (CPMs) traditionally requires iterative collaboration between clinical experts, data scientists, and informaticists. This process refines the often small but critical detail...
Programming over Thinking: Efficient and Robust Multi-Constraint Planning
arXiv:2601.09097v1 Announce Type: new
Abstract: Multi-constraint planning involves identifying, evaluating, and refining candidate plans while satisfying multiple, potentially conflicting constraints. Existing large language model (LLM) approaches face fundamental limitations in this domain. Pure r...
The Weaviate C# client is now generally available! This release brings a modern and intuitive API for .NET developers, making it easier than ever to build AI-powered applications.
Musk denies awareness of Grok sexual underage images as California AG launches probe
The California Attorney General has opened a formal investigation into Elon Musk's xAI after its chatbot Grok began generating nonconsensual sexual images of real women and even children.
Generative AI tool helps 3D print personal items that sustain daily use
“MechStyle” allows users to personalize 3D models, while ensuring they’re physically viable after fabrication, producing unique personal items and assistive technology.
The multi-billion AI security problem enterprises can’t ignore
AI agents are supposed to make work easier. But they’re also creating a whole new category of security nightmares. As companies deploy AI-powered chatbots, agents, and copilots across their operations, they’re facing a new risk: how do you let employees and AI agents use powerful AI tools without a...
In this post, we’ll explore when multi-agent architectures become necessary, the four main patterns we’ve observed, and how LangChain empowers you to effectively build multi-agent systems.
n8n has set out itself as one of the best low-code AI development platforms. The characteristic drag-and-drop interface of n8n has won the hearts of many coders and non-coders alike. The low entry barrier and high skill ceiling makes it the perfect tool for executing ideas on the go. But there is s...
Company raises $250K pre-seed led by Squared Circle Ventures, secures $100K in AWS credits, joins NVIDIA Inception, and drives 264 early-access registrations following ISC2 Security Congress keynote debut Assail, Inc., a cybersecurity company building autonomous AI agents for API-first offensive sec...
Aviatrix Introduces Two Zero Trust Security Programs
Aviatrix Breach Lock and the Aviatrix Threat Research Center Bolster Aviatrix Cloud Native Security Fabric and Help Organizations Stop Advanced Cloud Threats Aviatrix® today announced the launch of two new initiatives aligned to its Zero Trust for Workloads product: Aviatrix Breach Lock, a free rapi...
How AI Models Remember Things They Were Told to Forget
AI systems are often described as probabilistic, fuzzy, or approximate. That language makes forgetting sound natural. If a model is not deterministic, surely it cannot remember specific things. In practice, the opposite is true. AI models remember very well. They just remember differently than human...