AIQB
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 175 / 244 页 · 每页 40 stories

TutorialsOrdinary
Open-Source Jev Alternatives: Run Typed, Calibrated LLM Decisions Locally
TypeSafe's Jev put a name on something real: "System One" models that return typed, calibrated...
Source: DEV Community
TutorialsOrdinary
AI Built My Site in an Hour, Why Are You Charging So Much? — 4 Questions That Usually End the Conversation
I've had this conversation more times this year than in the previous five combined. A client sends...
Source: DEV Community
TutorialsOrdinary
I Built Shree AI OS — A Deterministic AI Runtime for Java 21
I Built Shree AI OS — A Deterministic AI Runtime for Java 21 After several months of...
Source: DEV Community
ResearchOrdinary
Blind Thermodynamic Ontology Discovery from Anonymous Experiments
Before a machine learning model can learn a thermodynamic equation of state, it must discover what its measurements represent: which channels scale with system size, which are intensive conjugates, how sectors pair through contact, and whi…
Source: arXiv
ResearchOrdinary
Leveraging Industrial Foundation Models at the Edge of Particle Physics Detectors via Distillation Learning and Hardware Co-design
Data acquisition (DAQ) systems at future particle physics experiments stand to benefit from the extremes of AI/ML development: large-scale foundation models can enhance the performance of feature extraction algorithms, and small-scale on-d…
Source: arXiv
ResearchOrdinary
Bayesian Filtering in Physical Systems via Test-time Trained Flow Matching
Bayesian filtering provides a principled framework for online state estimation under uncertainty, yet its application to systems with high-dimensional states and complicated posterior distributions remains challenging. Recent generative mo…
Source: arXiv
ResearchOrdinary
Discovering Physical Representation Languages
Before a machine can discover a physical law, it must discover what its measurements are: which observations live on cells, which are intensive or extensive, which sectors are dual, and which distinctions are merely gauge. We introduce phy…
Source: arXiv
TutorialsOrdinary
Orca: The Agent Development Environment for Running AI Coding Agents in Parallel
If you use Claude Code, Codex, or another AI coding agent in your terminal, you have probably hit...
Source: DEV Community · Claude
TutorialsOrdinary
AI Integration Cost: 2026 Enterprise Budgeting Guide
Determining the true AI integration cost is a major financial step for UK businesses looking to...
Source: DEV Community
ResearchOrdinary
Leaky-integrator reconstruction: taming error accumulation in recursive differenced time-series forecasting
We introduce leaky-integrator reconstruction, a training-free method that cures the error accumulation of recursive differenced forecasting. Our first contribution is diagnostic: predicting one-step changes and integrating them by cumulati…
Source: arXiv
ResearchOrdinary
One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents
Repository-level software engineering (SWE) comprises heterogeneous task categories, whose progress under pooled agentic reinforcement learning can be uneven: gains in some categories coincide with regressions in others, while aggregate re…
Source: arXiv
ResearchOrdinary
If You Hear It, Help Find It: Orthogonal Knowledge Distillation for Open-Vocabulary Audio-Visual Event Localization
Open-vocabulary audio-visual event localization (OV-AVEL) grounds a text-queried event in time from video, audio, and language. The supervision sources available to this task can differ in temporal-boundary reliability: on OV-AVEBench, our…
Source: arXiv
TutorialsOrdinary
Running Flux locally on a Mac: install, commands and two models compared
Building the engine, downloading 17 GB of weights, every flag explained, and two Flux models measured on one machine: one paints from scratch, the other edits a finished frame.
Source: DEV Community
TutorialsOrdinary
How I Structure MiniMax H3 Max Image-to-Video Prompts with Sound
Turning a still image into a video is already a strange creative problem. Adding sound makes it...
Source: DEV Community
TutorialsOrdinary
Gemini Hacked Three Companies Using the Dumbest Trick in the Book
You probably saw the headline this week: Google's AI autonomously hacked three companies. Cue the...
Source: DEV Community · Gemini
ResearchOrdinary
The Evidence Ladder for Reinforcement Learning in Healthcare: From Retrospective Policies to Trusted Interventions
Reinforcement learning (RL) offers a natural language for healthcare decisions whose conse- quences unfold over time, yet most reported progress remains far from routine intervention. Ex- isting surveys organize the field by algorithm or c…
Source: arXiv
ResearchOrdinary
Machine-Interpretable Information: Compiling Documents into Searchable and Readable Protocol States
Long-context language models interface with external knowledge through raw natural language. In retrieval-augmented systems, this creates a persistent index-payload schism: dense vectors enable searchable routing, but models must re-ingest…
Source: arXiv · GPT / Qwen / Llama
TutorialsOrdinary
OpenAI's GPT-5.6 Sol Sets New Record: Sub-100ms Response Time Changes Everything
The Numbers That Matter OpenAI just shipped GPT-5.6 Sol with what might be the most...
Source: DEV Community · OpenAI / GPT
TutorialsOrdinary
Anthropic, OpenAI Agents Caught Creating Fake Identities During Security Tests
The Report That Should Keep You Up at Night The UK's AI Security Institute (AISI) ran a...
Source: DEV Community · OpenAI / Claude
ResearchOrdinary
LLM-Based FORM Code Generation with Verification-Driven Fine-Tuning
FORM is a domain-specific symbolic manipulation language widely used in particle physics for processing the very large algebraic expressions arising from multi-loop Feynman diagram calculations. Despite its central role in precision theore…
Source: arXiv · Qwen
TutorialsOrdinary
From Zero to One: How I Built a Multi-Market B2B Badge & Trophy Site with an AI Agent Configurator
The Niche: Badges, Medals, Coins, and Trophies Most e-commerce builders chase apparel and...
Source: DEV Community
ResearchOrdinary
What Can a Recurrent State Safely Forget?
Recurrent models must preserve information that changes future behavior while suppressing hidden-state error. These objectives conflict: contraction improves stability, but contraction along a future-distinguishing direction destroys memor…
Source: arXiv
ResearchOrdinary
TicTacBench: Benchmarking Timing Closure Capabilities of Coding Agents
Recent advances in large language models (LLMs) have led to the emergence of coding agents capable of performing complex engineering tasks, including register-transfer level (RTL) design and optimization. Existing RTL benchmarks mainly eva…
Source: arXiv
TutorialsOrdinary
Under the Hood of Browser-use (100k+ Stars): DOM Tree Distillation, Vision Grounding, and Production Web Agents
Why do traditional Selenium and Playwright scripts constantly break against modern web frontends? We dissect the architecture of Browser-use, the breakout open-source project with over 100k GitHub stars: DOM tree pruning, Set-of-Mark (SoM)…
Source: DEV Community
TutorialsOrdinary
Hybrid-precision attention reduces compute cost with minimal accuracy loss
Mixed‑precision quantization can halve the compute cost of LLM attention while keeping accuracy loss...
Source: DEV Community
ResearchOrdinary
Human-guided physics-constrained AI agents construct an auditable model of soil-plug evolution
Engineering predictions require physical mechanisms to be translated consistently into equations, discretization, code, and validation, yet errors can propagate despite local checks. Artificial-intelligence (AI) agents automate scientific …
Source: arXiv
TutorialsOrdinary
This Week in AI: Anthropic's CEO Asked to Slow Down, OpenAI Disclosed 6 Incidents, and Developers Shipped More Agents
Dario Amodei's "We Must Pace the Frontier" essay, OpenAI's misalignment incident reports, the...
Source: DEV Community · OpenAI / Claude
ResearchOrdinary
Low-Rank Frequency Convolution and Noise-Range Augmentation for Real-Time Pitch Estimation on Edge Devices
Pitch estimation on an edge device is constrained in three ways at once. The model must be small, it must stay accurate when the input is noisy, and one frame must be produced inside the frame period. In this report the Frequency Convoluti…
Source: arXiv
ResearchOrdinary
Stochastic Reconfiguration as Statistical Filtering for Overparameterized Neural Quantum States
Stochastic reconfiguration (SR) is the standard optimizer for neural quantum states (NQS), but modern NQS often have far more parameters than Monte Carlo samples. We show that in this regime the diagonal shift is more than a numerical stab…
Source: arXiv
ResearchOrdinary
A Patient World Model for Early Forecasting of Digital Health Campaign Outcomes: Capabilities and Limits
Digital direct-to-consumer (DTC) health campaigns are usually measured after the fact. In-flight forecasting commonly relies on a separate classifier for every cutoff and horizon. We treat this task as a dynamic-system problem and build a …
Source: arXiv
ResearchOrdinary
Rethinking Class Imbalance for Single-Cell Foundation Models: A Systematic Benchmark Across Architectures and Long-Tail Loss Functions
Single-cell foundation models (scGPT, scBERT, Geneformer) achieve cell-type classification accuracy up to 97.5% in our experiments, yet this aggregate accuracy can mask systematic failure on rare, often disease-relevant cell populations th…
Source: arXiv
ResearchOrdinary
Co-occurrence Patterns of LoRA Adapters in Production Diffusion Model Inference Services
Low-rank adaptation (LoRA) has become a key technology for serving large-scale personalized large language models and diffusion models in the cloud. However, the co-occurrence patterns, resource contention relationships, and evolutionary r…
Source: arXiv
ResearchOrdinary
CSC: Calibrated Simplicity for Conflict-Aware Social Bot Detection in the LLM Era
Social bot detection is essential for protecting online platforms from misinformation amplification, coordinated manipulation, and distorted public discourse. However, large language models have made social bots much harder to detect from …
Source: arXiv
TutorialsOrdinary
The Architecture of Averages
A payments product begins by asking what a customer’s balance is, and ordinary CRUD answers that well...
Source: DEV Community
TutorialsOrdinary
How Image Translation Actually Works: From Pixels to Translated Text
Translating text on a web page is easy. The browser already knows where the text is, what each...
Source: DEV Community
TutorialsOrdinary
Test-Time Compute and GRPO in Practice: From PPO to Critic-Free Reinforcement Learning
A deep dive into the paradigm shift from pre-training scaling laws to test-time compute. We deconstruct the mathematical derivation of DeepSeek-R1's Group Relative Policy Optimization (GRPO), critic-free architecture advantages, emergent s…
Source: DEV Community · DeepSeek
TutorialsOrdinary
AI Didn't Hack OpenAI. A Missed Debian Backport and an SSO Misconfiguration Did
The headline that circulated this week sounds like a movie trailer: hackers breached OpenAI's...
Source: DEV Community · OpenAI
ResearchOrdinary
Graph Memory for LLM Agents: At What Cost? A Comparative Evaluation of Query, Ingest, and Update Performance Across Graph Database Engines
Graph databases are frequently positioned as categorically necessary for connected-data workloads, yet the systems dimension along which they actually differ - query planning, indexing, and data-readiness cost - is rarely isolated from ven…
Source: arXiv
TutorialsOrdinary
I found an agent skill that harvests browser credentials. It had 60,000 stars.
What checking the scripts — not the README — revealed about a popular research skill, and the...
Source: DEV Community
TutorialsOrdinary
I let an LLM judge inside Python's if statements, then ran Ansible's own tests on it
fuzzyif puts a plain-language question in an if condition. Replacing 418 lines of if/elif in Ansible's distribution detection with two calls, scored by Ansible's 90 official fixtures.
Source: DEV Community