May 2026

AI agents are reading executive email at Fortune 500 companies. They are moving stablecoins between corporate treasury wallets. They are closing six-figure procurement contracts on behalf of teams that have never seen the agent. They have been doing this for eighteen months, ever since Anthropic shipped Computer Use in October 2024 and OpenAI released Operator three months later. None of them have a verifiable identity. None of them have scoped authority that can be cryptographically verified. None of […]

Ver mais

Release: datasette-ip-rate-limit 0.1a0 The datasette.io site was being hammered by poorly-behaved crawlers, so I had Codex (GPT-5.5 xhigh) build a configurable rate limiting plugin to block IPs that were hammering specific areas of the site too quickly. Here’s the production configuration I’m using on that site for the new plugin: datasette-ip-rate-limit: header: Fly-Client-IP max_keys: 10000 exempt_paths: – “/static/*” – “/-/turnstile*” rules: – name: public-demo-multi-parameter-table-pages paths: – “/global-power-plants/*” – “/legislators/*” window_seconds: 60 max_requests: 60 block_seconds: 20 Tags: datasette, rate-limiting, […]

Ver mais

arXiv:2605.12908v1 Announce Type: new Abstract: Weak-to-strong (W2S) generalization, in which a strong model is fine-tuned on outputs of a weaker, task-specialized model, has been proposed as an approach to aligning superhuman AI systems. Existing theoretical analyses either fix the student’s representations or operate in restricted settings. Whether multi-step SGD can succeed in feature learning while preserving diverse pre-trained capabilities remains open. We study W2S in the setting of reward-model learning with two-layer neural networks. The strong model has […]

Ver mais

arXiv:2605.12899v1 Announce Type: new Abstract: Experimental design has emerged as a powerful approach for improving the sample efficiency of A/B testing, yet existing designs rely critically on correctly specified models. We study robust sequential experimental design under model misspecification and develop a unified framework that covers both contextual bandit and dynamic settings. Theoretically, we prove that our design bounds the worst-case mean squared error of the estimated treatment effect. Empirically, we demonstrate the effectiveness of the proposed approach […]

Ver mais

arXiv:2605.12768v1 Announce Type: new Abstract: Open time-series forecasting (TSF) benchmarks cover retail, energy, weather, and traffic, but supply-chain logistics remains underserved. We introduce ISOMORPH, the first public digital twin of a multi-echelon logistics network with fully interpretable, user-configurable parameters and modular topology, demand process, and control rules. The simulator advances a directed routing graph in discrete time: demand arrives at the destination, is served from stock or recorded as backlog, and triggers replenishment through the network. The state […]

Ver mais

arXiv:2605.12697v1 Announce Type: new Abstract: Length-dependent logit rescaling is widely used to stabilize long-context self-attention, but existing analyses and methods suggest conflicting inverse-temperature laws for the context length $n$, ranging from $(log n)^{1/2}$ to $log n$ and $(log n)^2$. We provide a general theory showing that the desirable scale is determined by the gap-counting function $N_n$ of each attention row. Counting how many competitors lie within each gap from the maximum, we define an upper-tail accumulation scale and […]

Ver mais

arXiv:2605.12668v1 Announce Type: new Abstract: Conformal prediction provides a principled framework for uncertainty quantification with finite-sample coverage guarantees. While recent work has extended conformal prediction to online and sequential settings, existing methods typically focus on a single coverage level and do not ensure consistency across multiple confidence levels. In many real-world applications, such as weather forecasting, macroeconomic prediction, and risk management, different users operate under heterogeneous risk tolerances and require calibrated uncertainty estimates across a range of coverage […]

Ver mais

arXiv:2605.12508v1 Announce Type: new Abstract: On-chain lending has expanded across multiple distributed ledgers as DeFi becomes increasingly multi-chain. This environment introduces novel technical and financial mechanisms, particularly cross-blockchain communication and asset transfer protocols, yet cross-chain elements remain understudied in lending protocol risk management. To address this gap, we applied panel regression fixed effects and OLS models to empirically analyze cross-blockchain interoperability solutions, using TVL and total revenue as performance proxies from October 2022 to January 2025. Our data […]

Ver mais