AI Intelligence Archive

On September 15 at 10:27 UTC, an agent working for me ran a dry-run seven times, once per language of...

Large language models (LLMs) on personal devices keep user data local, but limited memory restricts their general capability. Since most user-specific tasks do not require the breadth of frontier models, a practical solution is to adapt a …

Expert offloading enables large mixture-of-experts (MoE) language models to run locally, but conventional batching stalls the GPU while missing experts are fetched. A popular remedy, known as prefetching, is to predict what experts will be…

Reinforcement learning objectives that maximize expected return can concentrate probability mass on a small subset of high-reward outcomes, limiting diversity when multiple distinct solutions are desirable. We introduce Density-Aware Rewar…

Reinforcement learning with verifiable rewards (RLVR) has become an effective approach for improving mathematical reasoning in large language models. However, a fundamental question remains: when a model fails to solve a problem, does it t…

Large language models are candidates for collecting and analyzing records in criminal investigations and mass surveillance. Their capability and accuracy in this work cannot currently be measured. Real deployments lack the ground truth nee…

SAE-based feature-level interventions to suppress PII leakage in LLMs are primarily evaluated by log-probability reduction. We show that this criterion cannot distinguish feature specific suppression from a magnitude artifact: a direction-…

In single-pixel (ghost) imaging, known illumination patterns probe an object and a detector with no spatial resolution returns one number per pattern. Each measurement costs acquisition time and photons and, for ionizing probes, contribute…

Split Inference (SI) addresses the hardware resource limitations of Internet of Things (IoT) devices by distributing tiny Machine Learning (TinyML) model segments across multiple IoT devices. However, a device failure can interrupt the inf…

We study cooperative multi-agent reinforcement learning with decoupled dynamics and a monotone submodular reward function over joint trajectories. Due to the non-additive reward, an optimal policy may depend on the history of all agents. T…

A deployed LLM coaching agent answers requests whose correct response depends on player state and game version, so advice that is right for one player can be wrong for the next. We describe TactiNavi, the serving system behind the LLM coac…

Probabilistic time-series forecasting requires models that simultaneously capture structured temporal dynamics, expressive uncertainty, and efficient inference, yet existing approaches fall short of this goal: latent dynamical models impos…

An activation steering edit should depend on the input exactly when the destination varies \emph{inside} the class a query requests, and that is measurable before any rule is fitted or applied. Contrastive Activation Addition (CAA) adds on…

Your next customer doesn't scroll Retailers have a habit called a store walk. The manager walks the...

Hybrid models let neural networks supply the unresolved parameters of a process-based simulator, and are fitted to observations that constrain aggregate fluxes and storages but not the internal partitioning that produces them. Such a fit i…

Graph neural network explanations are often evaluated via deletion fidelity, i.e., by deleting their top-ranked edges and measuring the resulting change in model output. Yet, in a message-passing GNN, "deleting an edge" does not specify a …

Transaction event sequences support diverse prediction tasks in online commerce and digital financial services. The same transaction records can yield prediction targets with different semantics, including values, changes, and relations to…

A 2025 emoji shifted Claude Code's TUI one cell until a width-table update. How wcwidth, VS16, ZWJ and mode 2027 decide where your agent's borders land.

I’ve worked in Sydney restaurants and bars for over a decade, with every kind of booking app open at...

Stop building dashboards that look impressive but leave your team asking, “So, what should we do...

Most teams pick one model and send it everything. That is the expensive mistake. The question is not...

MCP exploded: by mid-2026 there are 10,000+ public servers. But trust in the ecosystem is name-based,...

Small multirotor drones produce a characteristic acoustic signature, and microphones offer a low-cost way to detect them. High accuracies have been reported on public drone audio benchmarks, often with random clip-level splits of datasets …

Recent advances in video modeling have driven substantial progress in world action models (WAMs). Existing approaches exploit pretrained video priors by decoding actions from latent representations or inferring them from generated videos. …

Multimodal perception is central to many computer vision applications, including robotics and autonomous driving, where different sensing modalities provide complementary cues. Self-supervised learning offers a scalable solution for exploi…

I like Screen Studio-style videos: a smooth cursor, zooms that follow the action, a nice gradient...

Robotics and AI systems are increasingly deploying large vision-language models (LVLMs) to follow high-level instructions, from routine household chores to more complex organizational tasks. Yet, broad directives like "clean the room" or "…

Abuse of a normal endpoint costs you capacity. Abuse of an endpoint that calls a model costs you money per request while everything looks healthy. Four controls, and the quota bug that charges you for every rejection.

Distributed learning has drawn considerable attention due to its capability of learning from distributed data without aggregating raw observations. However, frequent exchange of model updates can become the main bottleneck of communication…

Open your server log and you find names like ClaudeBot, GPTBot, ChatGPT-User, OAI-SearchBot and...

People don't think about loans in form fields. They think in sentences: "What's my EMI on a 20 lakh...

I run four small single-purpose sites: a Discord timestamp generator, a zalgo text generator, an...

I explored why defining input and output schemas is not enough to make AI tool calling reliable. But...

Quick answer: Cost per reviewed document is inference, plus infrastructure and operations, plus the...

Show the agent the misaligned button instead of describing it, then make a Playwright test and a pixel diff decide whether it's fixed.

TikTok Shop just made box-free, label-free returns easier through Happy Returns while AI screens...

TL;DR: Most stalled ML projects aren’t stuck on modeling. They’re stuck on undefined success...

Server software grew up around one assumption: the machine stays awake. People post screenshots of...

Git trusts cached stat data to skip reading files. An edit in the same second as an index write can look clean. Here is how Git catches it, with a script.

The 146-second wait in my first RTX Spark post was the cost of a brand-new 100K-token prompt. A...