Article system

AI Intelligence Archive

浏览 AIQB 已保存的 9747 stories AI intelligenceArticles，覆盖Models、Products、Industry、Research、教程与观点方法六大分类。

Indexed articles9747

Categories6

Selected stories2

All articles

Section 74 / 244 页 · 每页 40 stories

PerspectivesOrdinary2026-10-20 06:31

https://x.com/emollick/status/2112310600806072387

Source: X：Ethan Mollick (@emollick)

TutorialsOrdinary2026-10-20 06:11

The Fenced Block Principle

The problem: code that reads well but means something else If you vibe code — or supervise...

Source: DEV Community

TutorialsOrdinary2026-10-20 05:52

Last week in Agent Security Update 3: That's not Aoki!

An apology to start off with! I was too busy having fun in Sydney last week hacking with customers,...

Source: DEV Community

TutorialsOrdinary2026-10-20 05:40

A bunch of Vibe Coding idiosyncrasies to beginners

Do you know vibe coding? Before you answer, please open 20 PRs with AI, get burned at least once,...

Source: DEV Community

TutorialsOrdinary2026-10-20 05:34

All Nine of My Voice Agents Are One Giant Prompt. Here Is Where That Stops Working.

This week I went through every voice agent config I have exported on disk. Nine client agents: HVAC,...

Source: DEV Community

TutorialsOrdinary2026-10-20 05:16

Your AI backfill passed the quality check. More than a third of it was wrong.

Glossary Task (Snowflake): a scheduled SQL statement or procedure call. Created...

Source: DEV Community

TutorialsOrdinary2026-10-20 05:02

I Let an AI Agent Run Parts of My Real Businesses for a Year. Here Are the Receipts.

One year ago, I opened a terminal on a small home server and gave an AI agent its first real...

Source: DEV Community

TutorialsOrdinary2026-10-20 05:01

The Nx Malware Ran Claude Code With --dangerously-skip-permissions. Could It on Yours?

In 2025 the s1ngularity payload launched Claude Code, Gemini CLI and Amazon Q with approval prompts off. How it worked, and the settings that refuse it.

Source: DEV Community · Claude / Gemini

TutorialsOrdinary2026-10-20 04:34

Your vision model cannot see a broken layout. Measure the pixels instead.

A vision model said the layout was fine while the footer was missing. How a pixel-level gate replaced it: per-line margins, CDP screenshots and a control that actually fails.

Source: DEV Community

TutorialsOrdinary2026-10-20 04:33

How I Built a Business Solving Apple's Missing AirTag Feature: AirTag Location History

Originally published on my Substack. This week I launched airtaghistory.com. It keeps the AirTag...

Source: DEV Community

TutorialsOrdinary2026-10-20 04:14

Expose a React Workflow to AI Agents with WebMCP

A person opens a trip draft and asks an assistant: Prepare my SFO to JFK trip for November 12–15,...

Source: DEV Community

TutorialsOrdinary2026-10-20 04:07

Beyond Tokens: Can AI Agents Learn to Communicate Through Latent Representations?

Exploring communication efficiency in cooperative artificial intelligence. Introduction:...

Source: DEV Community

TutorialsOrdinary2026-10-20 04:00

Four models, one job each: what our own logs say about splitting the work

Most people pick one model and ask it to do everything. We stopped doing that months ago, mostly...

Source: DEV Community

TutorialsOrdinary2026-10-20 03:25

Can This Discount Actually Make Money? Building HustleRadar with Gemma 2 and SerpApi

HustleRadar: A Human-in-the-Loop AI Workspace for Safer Resale Decisions What I...

Source: DEV Community

TutorialsOrdinary2026-10-20 03:22

The path to AI FDE #01: Anatomy of an LLM app request

I've spent 8+ years building React/TypeScript front ends. Now I'm learning applied AI in public,...

Source: DEV Community

TutorialsOrdinary2026-10-20 03:13

Developers Won't Benefit Most From AI. Business People Will.

We assumed AI would give developers an unfair advantage. But the people actually winning with AI aren't always the most technical ones.

Source: DEV Community

TutorialsOrdinary2026-10-20 03:08

PromptShield AI: Zero-Leak Local Redaction & Prompt Hardening Engine for Open-Weight Models

A local-first, zero-telemetry PII and secrets redaction proxy with bidirectional rehydration for open-weight models (Gemma, Llama, Mistral).

Source: DEV Community · Llama / Mistral

TutorialsOrdinary2026-10-20 03:05

AI Agents Broke Into 100+ Sites at About $50 Each. The Confirmed One Used Old PHP Flaws

Reuters says AI agents run from exposed open-source panels breached 100+ sites at about $50 each. The one break-in NorForge verified used old PHP flaws.

Source: DEV Community

TutorialsOrdinary2026-10-20 02:57

We tested every small model our camera agent uses on a 4-core CPU. The bigger model lost.

Tool calling, vision, speech to text, text to speech and search, measured on 4 vCPUs with no GPU. The harnesses are open so you can rerun them.

Source: DEV Community

TutorialsOrdinary2026-10-20 02:42

We Turned a Multi-Agent AI System Into a Pixel-Art Kitchen So Students Can Learn How It Works

Agentic AI is everywhere right now, and for a student just getting started it can look impossible....

Source: DEV Community

TutorialsOrdinary2026-10-20 02:34

I compared 8 memory layers for AI coding agents, then built a ninth anyway

Every new chat with my coding agent started the same way. Ten minutes of me typing some version of...

Source: DEV Community

TutorialsOrdinary2026-10-20 02:30

Google Gemini 4 Argon Brings Multi-Agent Blender Workflows to Antigravity

Google has rolled out Gemini 4 Argon, a model positioned for complex, multi-step work across the...

Source: DEV Community · Gemini

TutorialsOrdinary2026-10-20 02:15

MAI-Transcribe-2 Leads David AI’s 21-Language Speech Recognition Benchmark

Microsoft AI’s MAI-Transcribe-2 achieved the lowest transcription error rate in 18 of 21 languages in...

Source: DEV Community

TutorialsOrdinary2026-10-20 02:08

How Can Text Inside a Document Hijack an AI Agent?

A user opens an email assistant and asks it to summarise their unread messages. Somewhere in their...

Source: DEV Community

TutorialsOrdinary2026-10-20 02:06

Your grammar checker uploads every word you type. I built one that never leaves your Mac.

Open the network tab while you use most AI writing tools. Watch what happens as you type an email, a...

Source: DEV Community

ResearchOrdinary2026-10-20 01:59

Text-to-Image Generative Modeling Without Language Pretraining

Modern text-to-image (T2I) systems use internet-scale language pretraining for text-conditioned image generation. We ask whether language pretraining is necessary for strong T2I generation. We introduce Halcyon, a family of early-fusion mu…

Source: arXiv

ResearchOrdinary2026-10-20 01:59

OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory

Multimodal large language models (MLLMs) have made rapid progress on visual reasoning, yet many tasks require evidence that is difficult to obtain through internal reasoning alone: our controlled study on 22 benchmarks shows that simply in…

Source: arXiv · GPT / Qwen / Kimi

ResearchOrdinary2026-10-20 01:59

Can It Actually Work? Synthesizing Functional CAD Mechanisms

While LLM agents can generate visually plausible CAD assemblies, these designs often fail to fulfill their intended physical functions. We study functional CAD design, a setting where agents interpret underspecified requests to construct a…

Source: arXiv

ResearchOrdinary2026-10-20 01:59

Reinforcement Learning with Visually Attributable On-Policy Self-Distillation

Multimodal reinforcement learning with verifiable rewards (RLVR) provides reliable task-level supervision, yet offers limited guidance for localizing errors in visual reasoning. On-policy self-distillation (OPSD) complements this supervisi…

Source: arXiv

ResearchOrdinary2026-10-20 01:59

CoopUQ: Robust Multi-Robot Social Navigation via Uncertainty-Aware Cooperative Forecasting

Reinforcement learning enables mobile robots to make fast and scalable navigation decisions in constrained human-shared environments. However, directly extending single-robot approaches to multi-robot settings often fails to ensure safety …

Source: arXiv

ResearchOrdinary2026-10-20 01:59

Where is Physical RSI capability from? An Observational Scaling Law of LLM-as-policy

In this paper, we systematically investigate the mechanisms, scaling behavior, and generalization of physical Recursive self-improvement. We introduce PhysRSI-eval to quantify physical RSI through a five-axis metric: First-Attempt Performa…

Source: arXiv

ResearchOrdinary2026-10-20 01:59

WeAct: Dynamic Weight Switching as Agent Actions

Long-horizon agents must learn heterogeneous capabilities from interleaved trajectories. Updating shared parameters to learn new capabilities can degrade previously acquired ones, resulting in catastrophic forgetting. As a trajectory unfol…

Source: arXiv · Qwen

ResearchOrdinary2026-10-20 01:59

Train An Imaginative Agent

A seemingly reasonable push in Sokoban can make the puzzle unsolvable. To avoid such mistakes, an agent has to anticipate the consequences of its actions before execution, especially when feedback from the real environment is sparse or act…

Source: arXiv

ResearchOrdinary2026-10-20 01:59

AnyGoal: On World Model Generalization Beyond Expert Rollouts

The goal of a world model is to capture the dynamics of an environment. This has the great appeal that it allows agents to plan and simulate a course of actions with the help of a world model. For these uses to be reliable, however, the le…

Source: arXiv

ResearchOrdinary2026-10-20 01:58

LatComm: Reusable Context Compression for Latent Communication between Large Language Models

Systems that combine several large language models (LLMs) often use one context for many questions. When the context is passed between models as text, the receiving model processes all of its tokens again for every question. Avoiding this …

Source: arXiv

ResearchOrdinary2026-10-20 01:58

Newton-Based Methods for Bilevel Optimization Beyond Lower-Level Strong Convexity

Bilevel optimization provides a framework for machine learning tasks such as hyperparameter optimization, data reweighting, and meta-learning. Existing Newton-based bilevel methods typically rely on the classical hypergradient formula, whi…

Source: arXiv · GPT

ResearchOrdinary2026-10-20 01:57

Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy

LLM agents that compute the right answer can still answer wrongly. Agent harnesses increasingly keep long tool outputs out of the context window, saving them to files that the agent searches or reads on demand. We ask where this costs accu…

Source: arXiv · GPT / Claude

ResearchOrdinary2026-10-20 01:56

Kronecker Delta Rule for Long Context Linear Attention

We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM family with a memory-revision mechanism that corrects targeted associat…

Source: arXiv · Hugging Face

ResearchOrdinary2026-10-20 01:56

Supervised Learning with Quantum Hamiltonian Embeddings

Machine learning is widely regarded as a promising application in which quantum computing could offer a computational advantage. However, general quantum machine learning models with provable guarantees on performance, trainability, and ef…

Source: arXiv

TutorialsOrdinary2026-10-20 01:56

How I built a multi-agent simulator as a native SwiftUI app

I built Sorcerer, a free, open-source macOS app (SwiftUI, macOS 15+, MIT) that simulates a community...

Source: DEV Community