arXiv:2602.18451v1 Announce Type: new Abstract: Contemporary science education reforms such as the Next Generation Science Standards (NGSS) demand assessments to understand students’ ability to use science knowledge to solve problems and design solutions. To elicit such higher-order ability, educators need performance-based assessments, which are challenging to develop. One solution that has been broadly adopted is Evidence-Centered Design (ECD), which emphasizes interconnected models of the learner, evidence, and tasks. Although ECD provides a framework to safeguard assessment validity, its […]

Ver mais

arXiv:2604.08720v1 Announce Type: new Abstract: Performance optimization of AI infrastructure is key to the fast adoption of large language models (LLMs). The PyTorch compiler (torch.compile), a core optimization tool for deep learning (DL) models (including LLMs), has received due attention. However, torch.compile is prone to correctness bugs, which cause incorrect outputs of compiled DL models without triggering exceptions, crashes, or warnings. These bugs pose a serious threat to the reliability of downstream LLM applications. Data from the PyTorch […]

Ver mais

arXiv:2602.01007v1 Announce Type: new Abstract: Byte Language Models (BLMs) have emerged as a promising direction for scaling language models beyond tokenization. However, existing BLMs typically require training from scratch on trillions of bytes, making them prohibitively expensive. In this paper, we propose an efficient distillation recipe that converts existing token-trained LLMs into BLMs while retaining comparable capabilities. Our recipe follows a two-stage curriculum: (1) Progressive Knowledge Distillation, which aligns byte-level representations with the embeddings of the token-trained teacher […]

Ver mais

Hi, I’ll be doing research at Mila Quebec this summer, and I’d love some advice on how to and what to prepare. The topic is Causal models for continual reinforcement learning. More specifically, the project hypothesizes that agents whose goal is to maximize empowerment gains will construct causal models of their actions and generalize better in agentic systems. For some background, I’m a last semester McGill undergraduate majoring in Statistics and Software Eng. I’ve done courses about: -PGMs: […]

Ver mais

arXiv:2602.13203v1 Announce Type: new Abstract: Telecommunication networks experience complex failures such as fiber cuts, traffic overloads, and cascading outages. Existing monitoring and digital twin systems are largely reactive, detecting failures only after service degradation occurs. We propose Adversarial Network Imagination, a closed-loop framework that integrates a Causal Large Language Model (LLM), a Knowledge Graph, and a Digital Twin to proactively generate, simulate, and evaluate adversarial network failures. The Causal LLM produces structured failure scenarios grounded in network dependencies […]

Ver mais

arXiv:2504.13336v2 Announce Type: replace Abstract: Flow Matching has recently gained attention in generative modeling as a simple and flexible alternative to diffusion models. While existing statistical guarantees adapt tools from the analysis of diffusion models, we take a different perspective by connecting Flow Matching to kernel density estimation. We first verify that the kernel density estimator matches the optimal rate of convergence in Wasserstein distance up to logarithmic factors, improving existing bounds for the Gaussian kernel. Based on […]

Ver mais

arXiv:2603.22248v1 Announce Type: cross Abstract: Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive (AR) models for language modeling, allowing flexible generation order and parallel generation of multiple tokens. However, this flexibility introduces a challenge absent in AR models: the emph{decoding strategy} — which determines the order and number of tokens generated at each iteration — critically affects sampling efficiency. Among decoding strategies explored in practice, confidence-based methods, which adaptively select which and how many […]

Ver mais