AI Intelligence Archive

In this paper, we study quantile distributional policy evaluation, where the return distribution under a given policy is approximated by the quantile Bellman fixed point $\bmη_m$. We investigate the statistical accuracy of estimating $\bmη…

The rapid expansion of the Internet of Things enables seamless collaboration among connected devices, making indoor localization a critical component of industrial automation. However, conventional localization techniques struggle in dynam…

Graph neural networks (GNNs) are among the standard tools for molecular property prediction, but the atom-level explanations of their predictions are hard for medicinal chemists to act on. We propose a change of input representation rather…

Despite the success of foundation models in image synthesis, achieving fine-grained spatial control remains a significant challenge, typically requiring auxiliary structural adapters or complex inference-time steering. While modern vision-…

Bayesian optimization (BO) is typically used only for problems where the underlying function is deemed "expensive" (e.g. in time or money), with the speed of the BO algorithm being of little importance. In the real world, however, problems…

Using a near-census of 802,863 resolved markets listed on Polymarket between January 2023 and May 2026, and only information available at listing (the market's title and description, named resolution source, and contract metadata), we pred…

Many learning systems are performative: once deployed, their predictions change the data distribution on which future models are trained. In multi-agent settings, this feedback is routed through a network of interacting learners, so stabil…

We consider online prediction from experts, a fundamental problem in machine learning, under differential privacy constraints. Existing private algorithms achieve near-optimal regret $\tilde{O}(\sqrt{T})$, where $T$ is the time horizon. Ho…

Graph-level out-of-distribution (OOD) generalization is critical for deploying graph neural networks in real-world applications where distribution shifts are inevitable. Existing approaches typically require class labels during representat…

Model predictive control is a high-performance control method, but it incurs comparatively high online computational costs across many control tasks and is challenging to realize in low-level hardware settings. To address this, we present …

One of the enduring challenges in online platforms is recommendation. Both communities and advertisers face difficulties in analyzing complex user behavior, identifying habits and evolving preferences, and providing timely recommendations …

Aligned large language models are trained to comply with benign requests while refusing harmful ones. However, recent work shows that this refusal behavior is fragile and can often be disrupted by simple activation- or weight-space interve…

Training modern AI models has become increasingly expensive, and updating deployed models can alter their internal feature representations, affecting applications built upon them. In retrieval systems, such updates often require complete g…

Linear-attention language models keep their context in a recurrent state of fixed size, where an inference system stores one copy of this state for every sequence it decodes at the same time. Storing the state at lower precision therefore …

Direct preference optimization (DPO) incurs training costs in output scoring, sequence execution, and backpropagation, so reducing supervision alone does not necessarily save computation. We introduce a unified framework for stochastic pru…

Autoregressive (AR) models have proved effective for zero-shot prediction from longitudinal electronic health records, but impose strictly sequential generation on trajectories containing temporally clustered and interdependent clinical ev…

We study an extremal regression problem, an example of multiple instance learning (MIL), where the predictors are given as multisets, called bags. The bag elements (called instances) possess hidden labels which are pooled to produce an obs…

Residual-based causal discovery assumes that each variable is generated from its parents and an independent noise source. It estimates this noise and tests its independence from candidate causes. For identifiable functions such as additive…

Masked language modelling (MLM) is the dominant pretraining objective for transformer encoders applied to longitudinal electronic health records (EHRs). However, reconstructing individual events may overemphasise noisy or weakly informativ…

Speculative decoding accelerates large language model inference by employing a compact draft model to propose candidate tokens that are verified in parallel by a target model. Conventional training relies on soft-label distillation, which …

The blog cost me zero dollars and it gets about a hundred thousand views a year. Both of those are...

As AI applications move from prototypes into production, managing model access becomes a different...

I run a reading site for language learners. Every line of every story has audio, generated with...

This is a submission for the Hacktoberfest Open-Source AI Challenge: Week 3. What I...

You can spot it in two seconds Open any landing page launched in the last year and give...

A step-by-step guide to writing a fixture and directive driven mock LLM provider that makes your tests and eval suites fast, free, and reproducible.

My coding agent has one rule it must never break: don't push straight to main, master or staging. The...

Hi, I'm Selçuk. I've been building an open-source tool called Kule, and its first preview release is...

Scribes, answer engines, chart copilots and patient assistants fail in different ways. What an engineer should check for each, from a year of building

OpenAI disclosed that two of its own coding agents weakened the monitoring built to catch their misbehavior — one suppressed security alerts, one deleted a safety monitor and then denied it. The three failure modes map one-to-one onto the …

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle ...

Production Kubernetes Operator guide: Kubebuilder architecture, CRD OpenAPI validation, Controller-Runtime reconciliation, and custom Go stateful controllers.

On the day we moved every business onto a new agent runtime, our Python service started growing in a...

AI infrastructure on Kubernetes: NVIDIA GPU Operator automation, MIG slicing, Kubelet Topology Manager NUMA alignment, Dynamic Resource Allocation, and KEDA scaling.

Automating software releases with CLI scripts works well until something goes wrong. A developer...

Invalid output, a 4,000 euro refund nobody approved, a provider outage and no trace of what happened. What a Java agent needs before production, with code.

"LLM-as-judge" has quietly become infrastructure. We use one model to grade another's answers, to...

Here's a question that sounds trivial and isn't: when your program produces an output, how do you...

Nine checks that decide whether an AI engine can quote your page, with the exact commands and the numbers measured on three live sites.

In my last post at Treat Your LLM Agent Like a New Bank Teller I argued that securing an LLM agent is...