arXiv:2604.14210v1 Announce Type: new Abstract: A claim has been circulating on social media and practitioner forums that Chinese prompts are more token-efficient than English for LLM coding tasks, potentially reducing costs by up to 40%. This claim has influenced developers to consider switching to Chinese for “vibe coding” to save on API costs. In this paper, we conduct a rigorous empirical study using SWE-bench Lite, a benchmark of software engineering tasks, to evaluate whether this claim of Chinese […]

Ver mais

arXiv:2602.02544v1 Announce Type: new Abstract: While Diffusion Language Models (DLMs) offer a flexible, arbitrary-order alternative to the autoregressive paradigm, their non-causal nature precludes standard KV caching, forcing costly hidden state recomputation at every decoding step. Existing DLM caching approaches reduce this cost by selective hidden state updates; however, they are still limited by (i) costly token-wise update identification heuristics and (ii) rigid, uniform budget allocation that fails to account for heterogeneous hidden state dynamics. To address these challenges, […]

Ver mais

arXiv:2603.19160v1 Announce Type: cross Abstract: Prediction-powered inference (PPI) is a rapidly growing framework for combining machine learning predictions with a small set of gold-standard labels to conduct valid statistical inference. In this article, I argue that the core estimators underlying PPI are equivalent to well-established estimators from the survey sampling literature dating back to the 1970s. Specifically, the PPI estimator for a population mean is algebraically equivalent to the difference estimator of Cassel et al. (1976), and PPI […]

Ver mais

Finding high-quality, cascading failure logs from real manufacturing to train continuous control RL agents is practically impossible due to proprietary air-gaps. Most open-source datasets are just Gaussian noise, which doesn’t respect the physical invariants needed for realistic state-transition dynamics. I’ve been experimenting with building a hybrid LLM-Physics simulation of a liquid-phase exothermic batch reactor to generate high-fidelity telemetry, and I’d love to get this community’s thoughts on the methodology for industrial environment design. **How we structured the state […]

Ver mais

arXiv:2603.15799v1 Announce Type: new Abstract: Prose2Policy (P2P) is a LLM-based practical tool that translates natural-language access control policies (NLACPs) into executable Rego code (the policy language of Open Policy Agent, OPA). It provides a modular, end-to-end pipeline that performs policy detection, component extraction, schema validation, linting, compilation, automatic test generation and execution. Prose2Policy is designed to bridge the gap between human-readable access requirements and machine-enforceable policy-as-code (PaC) while emphasizing deployment reliability and auditability. We evaluated Prose2Policy on the […]

Ver mais

Biased gradient compression with error feedback (EF) reduces communication in federated learning (FL), but under non-IID data, the residual error can decay slowly, causing gradient mismatch and stalled progress in the early rounds. We propose step-ahead partial error feedback (SA-PEF), which integrates step-ahead (SA) correction with partial error feedback (PEF). SA-PEF recovers EF when the step-ahead coefficient $α=0$ and step-ahead EF (SAEF) when $α=1$. For non-convex objectives and $δ$-contractive compressors, we establish a second-moment bound and a residual […]

Ver mais

arXiv:2602.11316v1 Announce Type: new Abstract: Prediction under uncertainty is a critical requirement for the deep neural network to succeed responsibly. This paper focuses on selective prediction, which allows DNNs to make informed decisions about when to predict or abstain based on the uncertainty level of their predictions. Current methods are either ad-hoc such as SelectiveNet, focusing on how to modify the network architecture or objective function, or post-hoc such as softmax response, achieving selective prediction through analyzing the […]

Ver mais

Hi all, I have trained a convolutional autoencoder on a set of medical images. Further classified latent feature maps using random forest to find the top scoring feature map. Now my goal is to understand which input image is captured in top scoring latent feature map. Any suggestions? I have tried encoding one image at a time while other images were muted. I then checked spearman between top scoring feature map with the original top scoring feature map. […]

Ver mais