Many reinforcement learning (RL) problems admit multiple terminal solutions of comparable quality, where the goal is not to identify a single optimum but to represent a diverse set of high-quality outcomes. Nevertheless, policies trained by standard expected return maximization routinely collapse onto a small subset of...
Post-training quantization (PTQ) methods for large language models compress pretrained weights to low bit-widths without retraining by optimizing layer-local proxy objectives. Activation-aware scaling methods such as AWQ protect input channels with large activation norms, whereas second-order error-compensation methods...
This thesis proposes a novel approach that uses Q-Learning on top of LGP, Reinforced Linear Genetic Programming (RLGP) to learn the optimal register-action assignments, and introduces a new framework "linear-gp" written in memory-safe Rust that allows for extensive experimentation for future works.
An accelerator-friendly QP solver is implemented that accumulates one Hessian per weight matrix and solves many small QPs in parallel with a convergence certificate per column, enabling one-shot post-training pruning at scale on a single accelerator without fine-tuning.
Mohammad Mozaffari, Samuel Kushnir, Maryam Mehri Dehnavi et al.· arXiv.org· 2 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Unequal representation of demographic groups in training data poses challenges to model generalisation across populations. Standard practice assumes that balancing subgroup representation always optimises performance. However, recent empirical results contradict this assumption: in some cases, imbalanced data distribut...
Anissa Alloula, Charles Jones, Zuzanna Wakefield-Skorniewska et al.· 0 citations
This work proposes AISTRAL, an AI-Native Internet architecture that supplies a native layer for publisher-controlled semantic discovery and retrieval, and proposes a web-native semantic resolver, analogous to DNS but operating over meaning rather than location.
Yi-Xi Chen, Muhammad Bilal, Tie Ma et al.· 1 citation
Hyperspectral image (HSI) classification typically involves large-scale data and computationally intensive training, which limits the practical deployment of deep learning models in real-world remote sensing tasks. This study introduces SpectralTrain, a universal, architecture-agnostic training framework that enhances...
Meihua Zhou, Liping Yu, Xinyu Tong et al.· 0 citations
Edge AI must adapt to new classes and shifting data within a tight power budget, but online continual learning (learning from each sample before the next arrives) is where conventional processors are least efficient: each small update pays a full dispatch at full power. We present CLP-SNN, a spiking architecture that r...
Elvin Hajizada, Danielle Rager, Timothy Shea et al.· 0 citations
We present a method that models the evolution of an unbounded number of time series clusters by switching among an unknown number of regimes with linear dynamics. We develop a Bayesian non-parametric approach using a hierarchical Dirichlet process as a prior on the parameters of a Switching Linear Dynamical System and...
Adri\'an P\'erez-Herrero, Paulo F\'elix, Jes\'us Presedo et al.· 0 citations
Large Language Models (LLMs) tend to generate a long reasoning chain when solving complex tasks. However, as the reasoning chain extends, critical intermediate steps and the original prompt will be buried in the context, receiving insufficient attention and leading to errors. In this work, we present ATAR, a novel reas...
The World-of-Whale benchmark (WoW-Bench) is introduced to evaluate low-level auditory perception and cognition using marine mammal vocalizations to better assess models' low-level listening and so that the models do not rely on prior knowledge of the sound events.
Jaeyeon Kim, Heeseung Yun, S. Woo et al.· Annual Meeting of the Associ...· 9 citations· ⚡1
This survey categorises six primary text-conditioned generative modalities and examines how four techniques, namely self-supervised learning, mixture of experts, reinforcement learning from human feedback, and chain-of-thought prompting, are adapted across them, distinguishing shared mechanisms from experimentally supp...
Longzhen Han, Awes Mubarak, Almas Baimagambetov et al.· 1 citation
