A Coding Guide to Demonstrate Targeted Data Poisoning Attacks in Deep Learning by Label Flipping on CIFAR-10 with PyTorch
The MarkTechPost tutorial walks readers through a targeted data‑poisoning experiment that flips a portion of CIFAR‑10 labels from a chosen class to a malicious class using PyTorch. By constructing parallel clean and poisoned training pipelines with a lightweight ResNet‑18, the guide isolates the effect of label corruption on model behavior. With a 40% poison ratio, the poisoned model exhibits systematic misclassification of the targeted class while overall accuracy stays relatively stable. Visual diagnostics such as confusion matrices and per‑class reports highlight the attack’s precision.
Meet SETA: Open Source Training Reinforcement Learning Environments for Terminal Agents with 400 Tasks and CAMEL Toolkit
Researchers from CAMEL AI, Eigent AI and partners released SETA, an open‑source stack that couples a terminal‑focused toolkit with 400 synthetic reinforcement‑learning tasks. The framework delivers state‑of‑the‑art results on the Terminal Bench benchmark, hitting 46.5% accuracy on version 2.0 with a...
How to Build Portable, In-Database Feature Engineering Pipelines with Ibis Using Lazy Python APIs and DuckDB Execution
The tutorial shows how Ibis can create a portable, in‑database feature‑engineering pipeline that feels like Pandas but runs entirely in DuckDB. By registering data in the backend and keeping all transformations lazy, the code is translated into efficient SQL without...
Stanford Researchers Build SleepFM Clinical: A Multimodal Sleep Foundation AI Model for 130+ Disease Prediction
Stanford Medicine researchers unveiled SleepFM Clinical, a multimodal foundation model trained on 585,000 hours of polysomnography from about 65,000 individuals. The model learns a unified representation of brain, heart, and respiratory signals and can predict long‑term risk for more than...
A Coding Implementation to Build a Unified Apache Beam Pipeline Demonstrating Batch and Stream Processing with Event-Time Windowing Using DirectRunner
The tutorial shows how to build a unified Apache Beam pipeline that can run in both batch and stream‑like modes using the DirectRunner. It creates synthetic event‑time data, applies fixed windows with triggers and allowed lateness, and demonstrates how Beam...
Liquid AI Releases LFM2.5: A Compact AI Model Family For Real On Device Agents
Liquid AI unveiled LFM2.5, a compact 1.2 billion‑parameter model family designed for on‑device and edge inference. The suite includes Base, Instruct, Japanese, vision‑language, and audio variants, all released with open weights on Hugging Face and via the LEAP platform. Pre‑training was...
LLM-Pruning Collection: A JAX Based Repo For Structured And Unstructured LLM Compression
Zlab Princeton has open‑sourced the LLM‑Pruning Collection, a JAX‑based repository that aggregates leading pruning techniques for large language models. The repo bundles block‑level, layer‑level, and weight‑level methods—including Minitron, ShortGPT, Wanda, SparseGPT, Magnitude, Sheared LLaMA and LLM‑Pruner—under a unified training and evaluation...
Tencent Researchers Release Tencent HY-MT1.5: A New Translation Models Featuring 1.8B and 7B Models Designed for Seamless On-Device and Cloud...
Tencent Hunyuan researchers unveiled HY-MT1.5, a bilingual translation family comprising a 1.8 B and a 7 B model. Both models cover 33 languages plus five dialect variants and are released with open weights on GitHub and Hugging Face. The compact 1.8 B variant runs...
AI Interview Series #5: Prompt Caching
Prompt caching reduces LLM API costs by reusing static prompt components. By storing key‑value attention states in GPU memory, identical prefixes avoid recomputation, cutting latency and token usage. Engineers can boost efficiency by analyzing request patterns, restructuring prompts so shared...
A Coding Implementation to Build a Self-Testing Agentic AI System Using Strands to Red-Team Tool-Using Agents and Enforce Safety at...
The tutorial builds a red‑team evaluation harness with Strands Agents to stress‑test a tool‑using AI assistant against prompt‑injection and tool‑misuse attacks. It defines a guarded target agent, a red‑team agent that auto‑generates adversarial prompts, and a judge agent that scores...
Tencent Released Tencent HY-Motion 1.0: A Billion-Parameter Text-to-Motion Model Built on the Diffusion Transformer (DiT) Architecture and Flow Matching
Tencent Hunyuan’s 3D Digital Human team launched HY‑Motion 1.0, an open‑weight text‑to‑3D human motion model built on a Diffusion Transformer (DiT) architecture and trained with Flow Matching. The flagship model contains 1 billion parameters, with a Lite 0.46 billion variant, and generates SMPL‑H...
A Coding Implementation of an OpenAI-Assisted Privacy-Preserving Federated Fraud Detection System From Scratch Using Lightweight PyTorch Simulations
The tutorial walks through building a privacy‑preserving federated fraud‑detection system from scratch using lightweight, CPU‑only PyTorch. It simulates ten independent banks, partitions highly imbalanced transaction data with a Dirichlet distribution, and coordinates local model updates via a FedAvg loop. After...
Meet LLMRouter: An Intelligent Routing System Designed to Optimize LLM Inference by Dynamically Selecting the Most Suitable Model for Each Query
LLMRouter, an open‑source library from UIUC, sits between applications and heterogeneous LLM pools to automatically select the most appropriate model per query. It offers over 16 routing algorithms organized into single‑round, multi‑round, personalized, and agentic families, each configurable via a...
NVIDIA AI Researchers Release NitroGen: An Open Vision Action Foundation Model For Generalist Gaming Agents
NVIDIA’s AI team unveiled NitroGen, an open‑source vision‑action foundation model that learns to play commercial games directly from pixel inputs and gamepad actions. The model is trained on 40,000 hours of filtered gameplay video spanning over 1,000 titles, using automatic...
Liquid AI’s LFM2-2.6B-Exp Uses Pure Reinforcement Learning RL And Dynamic Hybrid Reasoning To Tighten Small Model Behavior
Liquid AI released LFM2-2.6B-Exp, an experimental checkpoint that adds a pure reinforcement‑learning (RL) stage to its 2.6 billion‑parameter LFM2 model. The RL fine‑tuning targets instruction following, knowledge retrieval, and math without altering the hybrid convolution‑attention architecture. Benchmark results show the model...
How to Build Production-Grade Agentic Workflows with GraphBit Using Deterministic Tools, Validated Execution Graphs, and Optional LLM Orchestration
The tutorial demonstrates how to build a production‑grade, agentic workflow for customer‑support ticket triage using GraphBit. It starts by configuring the GraphBit runtime, defining typed ticket data, and registering deterministic tools for classification, routing, and response drafting. These tools are...
A Coding Implementation on Building Self-Organizing Zettelkasten Knowledge Graphs and Sleep-Consolidation Mechanisms
The tutorial by Asif Razzaq demonstrates how to build a self‑organizing Zettelkasten memory system for agentic AI using Google Gemini. It defines a MemoryNode data class, ingests text by atomizing it into discrete facts, embeds each fact, and links semantically...
A Coding Guide to Build an Autonomous Multi-Agent Logistics System with Route Planning, Dynamic Auctions, and Real-Time Visualization Using Graph-Based...
The tutorial walks readers through building a fully autonomous, multi‑agent logistics simulation where five smart trucks navigate a 30‑node graph‑based city. Each truck acts as an agent that bids on delivery orders, plans shortest‑path routes, monitors battery levels, and seeks...
This AI Paper From Stanford and Harvard Explains Why Most ‘Agentic AI’ Systems Feel Impressive in Demos and Then Completely...
The Stanford‑Harvard paper "Adaptation of Agentic AI" introduces a unified, mathematically grounded framework for tuning agentic AI systems that combine large language models with planning, tool‑use, and memory modules. It categorizes adaptation into four paradigms based on whether the target...
InstaDeep Introduces Nucleotide Transformer V3 (NTv3): A New Multi-Species Genomics Foundation Model, Designed for 1 Mb Context Lengths at Single-Nucleotide...
InstaDeep unveiled Nucleotide Transformer v3 (NTv3), a multi‑species genomics foundation model that processes 1 megabase windows at single‑nucleotide resolution. The U‑Net‑style architecture combines down‑sampling, transformer layers, and up‑sampling to deliver both prediction and controllable sequence generation. NTv3 was pre‑trained on 9 trillion base...
Google Health AI Releases MedASR: A Conformer Based Medical Speech to Text Model for Clinical Dictation
Google Health AI released MedASR, an open‑weights medical speech‑to‑text model built on the Conformer architecture. The 105‑million‑parameter model is trained on roughly 5,000 hours of de‑identified physician dictations and clinical conversations across radiology, internal and family medicine. Benchmarks show MedASR...
How to Build a Proactive Pre-Emptive Churn Prevention Agent with Intelligent Observation and Strategy Formation
The tutorial walks through building a Pre‑Emptive Churn Prevention Agent that automatically spots inactive users, evaluates their churn risk, and drafts personalized re‑engagement emails using Google Gemini. It creates a mock customer database, defines a risk‑analysis prompt, and generates incentive‑driven...
Google DeepMind Researchers Release Gemma Scope 2 as a Full Stack Interpretability Suite for Gemma 3 Models
Google DeepMind has unveiled Gemma Scope 2, an open‑source interpretability suite that spans the entire Gemma 3 family from 270 M to 27 B parameters. The platform leverages layer‑wise sparse autoencoders and transcoders, trained on roughly 110 petabytes of activation data and over a trillion...
How to Build a Fully Autonomous Local Fleet-Maintenance Analysis Agent Using SmolAgents and Qwen Model
The tutorial demonstrates how to build a fully autonomous fleet‑maintenance analysis agent using SmolAgents and a locally hosted Qwen 2.5‑7B‑Instruct model. By defining a custom tool that loads telemetry CSV files, the agent can reason step‑by‑step, flag trucks with high engine...
Anthropic AI Releases Bloom: An Open-Source Agentic Framework for Automated Behavioral Evaluations of Frontier AI Models
Anthropic has open‑sourced Bloom, an agentic framework that automates behavioral evaluations of frontier AI models. Researchers provide a seed file describing a target behavior, and Bloom generates dozens to hundreds of diverse, reproducible scenarios through a four‑stage pipeline of understanding,...
AI Interview Series #4: Explain KV Caching
KV caching is an inference optimization that stores the keys and values from previous attention steps during autoregressive generation. By reusing these cached tensors, the model only computes queries for new tokens, avoiding redundant attention over the entire prompt. Benchmarks...
Unsloth AI and NVIDIA Are Revolutionizing Local LLM Fine-Tuning: From RTX Desktops to DGX Spark
Unsloth, a GPU‑optimized fine‑tuning library, now runs on everything from GeForce RTX desktops to NVIDIA's DGX Spark, delivering up to 2.5× faster training for LLMs. The tool supports parameter‑efficient methods like LoRA/QLoRA, full model updates, and reinforcement‑learning pipelines, each with clear...
How to Orchestrate a Fully Autonomous Multi-Agent Research and Writing Pipeline Using CrewAI and Gemini for Real-Time Intelligent Collaboration
The tutorial demonstrates how to build a two‑agent CrewAI pipeline that leverages the Gemini Flash model for real‑time research and writing. It walks through environment setup, secure Gemini authentication, and the definition of a researcher and a writer agent with...
Thinking Machines Lab Makes Tinker Generally Available: Adds Kimi K2 Thinking And Qwen3-VL Vision Input
Thinking Machines Lab has moved its Tinker training API to general availability, removing the waitlist and opening access to all developers. The update adds three major capabilities: support for the 1‑trillion‑parameter Kimi K2 Thinking reasoning model, an OpenAI‑compatible sampling interface, and multimodal...
How to Design a Gemini-Powered Self-Correcting Multi-Agent AI System with Semantic Routing, Symbolic Guardrails, and Reflexive Orchestration
The article presents a step‑by‑step tutorial for building a Gemini‑powered multi‑agent AI system that uses semantic routing, symbolic guardrails, and a self‑correction loop. It defines a shared AgentMessage format, a CognitiveEngine that calls Gemini‑2.0‑Flash, and a SemanticRouter that maps user...
How to Design a Fully Local Agentic Storytelling Pipeline Using Griptape Workflows, Hugging Face Models, and Modular Creative Task Orchestration
The article walks through building a fully local, API‑free storytelling system using Griptape and a TinyLlama model from Hugging Face. It demonstrates an agent equipped with a calculator tool, hierarchical world‑generation and character tasks, and a final story‑writing task governed by...
CopilotKit v1.50 Brings AG-UI Agents Directly Into Your App With the New useAgent Hook
CopilotKit v1.50 rebuilds its frontend on the Agent User Interaction (AG‑UI) protocol and ships a new React hook, useAgent, that turns agent‑to‑UI communication into a single typed event stream. The hook subscribes to messages, streaming tokens, tool calls and shared state,...