The paper investigates when pooling information across intermittent-demand series is beneficial, focusing on the trade‑off between forgetting past data and borrowing strength from other series. By applying a single exponential recency operator to both item‑ and group‑level statistics in a hierarchical empirical‑Bayes hurdle model, the authors derive a diagnostic that quantifies credibility and resolution, and use it to screen for regimes where refinement offers no advantage. Empirical tests on five public panels of over 19,000 series show that the diagnostic accurately predicts when shared priors improve accuracy, and the resulting forecaster (EBB) matches specialist Gaussian‑process models within a few percent while being far cheaper to run.
arXiv:2608. 19888v1 Announce Type: cross Abstract: Streaming systems that maintain a pool of expert models must repeatedly decide whether to reuse an existing expert for arriving data, spawn a new one, or defer.
arXiv:2606. 03736v2 Announce Type: replace-cross Abstract: We study resource-constrained dynamic pricing when the seller seeks revenue and valid inference about demand at a price fixed before the selling season.
arXiv:2609.38914v1 Announce Type: new Abstract: Evaluating interactive agents is expensive. Agent behavior is stochastic, so reliability must be measured over repeated trials, but failures are rare a...
ERRAND is a new method for budgeted maintenance of agent memory that treats revalidation of stored knowledge as a priced errand competing for scarce actions. It uses an errand index that is single‑peaked, allowing certainty in either direction to cost nothing, and repairs by writing new versions rather than deleting old ones. In experiments across two drifting tool‑use worlds, ERRAND outperforms non‑oracle policies, achieving up to 10.0 percentage points improvement over eager revalidation while using only 11.0% of steps, and it self‑terminates when no budget is imposed.
The paper introduces a Causal Market Simulator using a continuous‑time causal prior‑data fitted network (PFN) to estimate market impact, addressing bias from informational flow that traditional models overlook. On a synthetic market prior with adjustable confounding, the PFN matches an unbiased oracle while a naive estimator overestimates by up to 68%. In zero‑shot tests on real limit‑order books and equity data, the model’s intervention responses align with ground‑truth scales and improve correlation with measured impact when a recurrent state‑space variant is used.
arXiv:2609.39634v1 Announce Type: cross Abstract: Common policy improvement methods, including TRPO, PPO, and GRPO, estimate policy improvement under the behavioral policy's state-visitation distribu...
arXiv:2607. 13618v1 Announce Type: new Abstract: LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed.
arXiv:2609. 20193v1 Announce Type: new Abstract: Retrieval plug-ins supply a deep forecaster with information its lookback window cannot carry.
The paper argues that large language model (LLM) providers, constrained by compute, often degrade service during congestion by routing queries to smaller models, cutting reasoning effort, or truncating context. It shows that this practice misrepresents costs because degraded answers can fail, leading to retries that inflate traffic or churn that erodes lifetime value. By modeling inference allocation with newsvendor, retry, and queueing frameworks, the authors derive a ‘shadow price of intelligence’ that quantifies the marginal value of each query, revealing that throttling under congestion acts as a demand lever rather than a cost lever.
LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the final cost cannot say why an agent failed: it may have misread the world, or read it correctly and still failed to act (the knowing-doing gap).
arXiv:2610.20104v1 Announce Type: new Abstract: Autoregressive transformers trained on limit order book (LOB) message streams are increasingly proposed as foundation models for market microstructure....
The paper proves that in time‑series validation three desirable properties—training sufficiency, test coverage, and temporal causality—cannot all be satisfied simultaneously. It introduces quantitative bounds involving the smallest training fraction (α), test coverage (β), future training fraction (Λ), and distance to nearest future training point (δ), showing that exceeding the causal frontier α+β=1 requires training on future data that must lie within (1−α)T of a test point. The authors demonstrate that the impact of such future leakage depends on distance rather than amount, and compare different validation schemes (walk‑forward, k‑fold, purged k‑fold) in terms of their position on this Pareto frontier, illustrating the trade‑offs with empirical results on noise data.