arXiv:2601.16120v1 Announce Type: new Abstract: Imbalanced classification, where one class is observed far less frequently than the other, often causes standard training procedures to prioritize the majority class and perform poorly on rare but important cases. A classic and widely used remedy is to augment the minority class with synthetic examples, but two basic questions remain under-resolved: when does synthetic augmentation actually help, and how many synthetic samples should be generated? We develop a unified statistical framework for […]

Ver mais

arXiv:2602.22282v1 Announce Type: cross Abstract: Data privacy is important in the AI era, and differential privacy (DP) is one of the golden solutions. However, DP is typically applicable only if data have a bounded underlying distribution. We address this limitation by leveraging second-moment information from a small amount of public data. We propose Public-moment-guided Truncation (PMT), which transforms private data using the public second-moment matrix and applies a principled truncation whose radius depends only on non-private quantities: data […]

Ver mais

We study the problem of online multi-class learning with partial feedback: in each trial of online learning, instead of providing the true class label for a given instance, the oracle will only reveal to the learner if the predicted class label is correct. We present a general framework for online multi-class learning with partial feedback that adapts the potential-based gradient descent approaches (Cesa-Bianchi & Lugosi, 2006). The generality of the proposed framework is verified by the fact that […]

Ver mais

ClearML’s 2025–2026 State of AI Infrastructure report reveals a costly disconnect in enterprise AI operations: nearly half of organizations waste millions on idle GPU capacity while AI teams wait in queues. Manual provisioning, weak automation, vendor lock-in, and immature governance block ROI just as AI agents and large-scale deployments accelerate.

Ver mais

arXiv:2601.09805v1 Announce Type: new Abstract: Modern logical reasoning with LLMs primarily relies on employing complex interactive frameworks that decompose the reasoning process into subtasks solved through carefully designed prompts or requiring external resources (e.g., symbolic solvers) to exploit their strong logical structures. While interactive approaches introduce additional overhead, hybrid approaches depend on external components, which limit their scalability. A non-interactive, end-to-end framework enables reasoning to emerge within the model itself — improving generalization while preserving analyzability without any […]

Ver mais

The idea of ​​”Privacy-Preserving AI” usually stops at local inference. You run a model on a phone, and the data stays there. But things get complicated when you need to prove to a third party that the output was actually generated by a specific, untampered model without revealing the input data. I’ve been looking into the recently open-sourced Remainder prover (the system Tools for Humanity uses for World). From an ML engineering perspective, the choice of a GKR […]

Ver mais

arXiv:2603.00549v1 Announce Type: new Abstract: We present PM2Lat, a fast and generalized framework for accurately predicting the latency of deep neural network models on GPUs, with special focus on NVIDIA. Unlike prior methods that rely on deep learning models or handcrafted heuristics, PM2Lat leverages the Single-Instruction-Multiple-Thread architecture of GPUs to model execution time of DNN models. First, we dive into fine-grained GPU operation modeling by studying computational behavior and memory access patterns. After identifying these characteristics, we found […]

Ver mais

arXiv:2602.16721v1 Announce Type: new Abstract: Numerous models have shown great success in the fields of speech recognition as well as speech synthesis, but models for speech to speech processing have not been heavily explored. We propose Speech to Speech Synthesis Network (STSSN), a model based on current state of the art systems that fuses the two disciplines in order to perform effective speech to speech style transfer for the purpose of voice impersonation. We show that our proposed […]

Ver mais