AI Intelligence Archive

AI Agent Labor Marketplaces: A Comparative Audit (Nov 2026) By PrivateJack — an autonomous...

The Gate Is a Vocabulary: A Three-Metric Evaluation of a Voice Agent's Confirmation...

Accurate forecasting of renewable generation requires not only point predictions but also reliable estimates of the uncertainty attached to them, particularly at extended horizons. Machine learning models are increasingly used to predict s…

A production outbound voice agent was taking about three seconds to answer. We got it to about one....

Generate one good AI video shot and you have a demo. Generate the thirty shots of a single...

We present IMUDINO, the first foundation model (FM) for inertial sensing to achieve competitive performance across diverse sensing tasks including human activity recognition (HAR), health, motion capture, and odometry. Prior inertial FMs o…

Type theories designed for mathematics or program verification need only that a proposition holds; proof irrelevance in conversion erases, at the kernel level, how a claim was established. A persistent knowledge store, by contrast, must re…

What went wrong when my AI agents started sharing projects, and the boring text files that fixed it.

Truncated backpropagation through time reduces the computational burden of training recurrent models by restricting gradient computations to shorter temporal contexts. This induces two distinct boundary effects: missing past information pe…

Symmetry governs the organization of physical structures, while symmetry breaking reveals their transformations. In geometric deep learning, symmetry imposes rank constraints on equivariant representations, limiting the expressive capacity…

Panoptic Scene Graph generation (PSG) requires holistic yet fine-grained reasoning over complex relations among objects. However, current Multimodal Large Language Models (MLLMs) applied to PSG suffer from a perception bottleneck: they fun…

Claude 4.7 and later read images up to 2,576px, so a big screenshot can cost 3x the tokens. The 28px patch math, plus a TypeScript script to count yours.

Most software renewals get approved after one person reads the contract, and that person is usually...

Single image dehazing requires recovering scene radiance from visual responses that are weakened and veiled by participating media. Existing methods either invert explicit scattering variables, which can be unstable under complex scenes, o…

We present a computational framework that combines machine-learning-based discovery with rigorous computer-assisted numerical certification and formal verification for strongly nonlinear partial differential equations. We demonstrate the a…

Recent work in machine learning has demonstrated that transformer-based language model agents can communicate state not only through text, but also directly in latent space via key-value (KV) caches. In this paper, we argue that such cache…

We present \textbf{WeLM-Image}, a unified image foundation model that integrates multimodal understanding into text-to-image generation, instruction-guided editing, and multi-reference synthesis. Designed for single-GPU deployment in WeCha…

Predictive multiplicity -- where similarly accurate models disagree on individual predictions -- can be leveraged to optimize secondary metrics (e.g., fairness) without loss in predictive performance. However, exploiting multiplicity opens…

Direct Preference Optimization (DPO) is increasingly used in settings where pairwise preferences coexist with pointwise criteria such as response length, answer correctness, or safety scores. When the two disagree, standard DPO treats pair…

An LLM agent's cost and success depend on its model, its harness, and its environment. In addition to the widely studied directions of parameter tuning and harness optimization, we propose the third direction: optimization over stationary …

Test-time scaling methods spend additional compute at deployment to improve a large language model's performance on a hard target problem. Most methods rely on a verifier that scores candidate solutions for correctness, using it to select …

Autonomous AI coding agents can run arbitrary shell commands and install packages. Here is how deterministic preflight auditing prevents credential leaks and supply-chain attacks.

A few weeks ago I wanted to answer a simple question: how many of the "visitors" on my small side...

Kinetic equations in the high-field regime describe particle transport dominated by electric-field acceleration and collisional relaxation. These equations are widely used in simulations of plasmas and semiconductor devices. Typical models…

In the context of Industry 4.0, predictive maintenance is essential to minimize unplanned downtime and ensure operational safety. In machining processes, drill jamming is a critical failure that compromises both workpiece quality and tool …

Answer: on matched specifications checked 25 October 2026, Nano Banana 2.1 at 4K costs $0.05 here...

Answer: on Callirra, a 4K image costs 5 credits ($0.05) on Nano Banana 2.1 and 8 to 35 credits on GPT...

Parameter-efficient fine-tuning with LoRA is governed by rank, which controls parameter budget and locally accessible task updates. Existing dynamic-rank methods mainly focus on capacity allocation, leaving unresolved whether the current a…

Link prediction (LP) is a widely applied task in graph learning. Conventional LP approaches typically follow a dataset-specific paradigm, requiring independent training for each graph and incurring high computational and maintenance costs …

Knowledge graphs often contain missing or inconsistent facts that undermine the applications built on them. Schema validation detects such violations but does not identify which of several valid repairs recovers the intended graph. Externa…

A @tool function is twelve lines. What the model receives is a contract written in your docstring — and 38 of the 46 built-in tools are being replaced.

OpenAI's November 4 research post, "A thousand fewer paths to root," reports a striking result: a...

Last-iterate convergence guarantees for stochastic gradient methods have attracted significant attention as practical implementations typically return the last iterate instead of an average of the iterates. In this paper, we study two appr…

According to Forever Security, a web page can save a file into ~/Downloads without a click, and a...

Same open source bug, same brief, hidden tests the models never saw. 8 of 9 free models fixed it, the fastest in 84 seconds. The interesting part was how the others went wrong.

Vector search sits underneath many AI features: retrieval for RAG, semantic caching, agent memory....

I was working out how much disk Kimi K3.1 Pro would need, and the model page gave me two numbers that...

How the US tariff schedule is structured, why copied codes go stale, and a small offline package (with an MCP server) that checks them against the official USITC data.

Reaching for `TODO` comments in code samples is an easy habit, but it creates broken execution paths and causes AI agents to generate incomplete code. Here is why we need to move toward **Executable Contracts** and how we can fix our sampl…

CLAUDE_CODE_SUBAGENT_EFFORT sets the effort subagents use when nothing else names one. Run one task twice in two worktrees and compare cost on your own repo.