Voice agents built on large language models (LLMs) now handle customer calls, but in regulated applications completing the task is not enough: the agent must follow a conversational flow set by law. In debt collection, it must confirm it is speaking to the debtor before mentioning the debt, identify itself and the cred...
Scaling up transformers has produced models with broad capabilities, deployed across heterogeneous hardware and fluctuating workloads. Yet, model families offer only a few fixed sizes (e.g., S, B, L), forcing users to commit to a model based on hardware and latency constraints. This rigidity is often suboptimal under v...
Walter Simoncini, Dheeraj Varghese, Gertjan J. Burghouts et al.· 0 citations
We introduce Continuum Attention (CA), a causal Transformer attention operator that replaces discrete key and value lookup with integration over a continuous, piecewise linear memory curve. Each query induces an exponential relevance density along the curve. Integrating the density and interpolated values yields exact...
Post-training binds the competence of an LLM agent to the interface it was trained on. Semantics-preserving perturbations of that interface, such as renaming an action or changing the notation, leave the underlying task unchanged yet make the agent worse at it. Scoring each sample independently means interface randomiz...
On-policy distillation (OPD) trains a student model on its own rollouts with token-level supervision from a teacher, assuming that the teacher can reach a correct answer from any prefix the student writes. We call this property the recoverability of the teacher and show it fails on incorrect rollouts: the teacher follo...
Kaixiang Zhao, Tianrun Yu, Shangzhe Li et al.· 0 citations
Potato is an important food-security crop in Ethiopia, but its production is affected by foliar diseases, particularly early blight and late blight. This study develops a lightweight convolutional neural network (CNN) to classify potato leaves into three categories: healthy, early blight, and late blight. The proposed...
Reinforcement learning with verifiable rewards (RLVR) trains language models from checked task outcomes, but assigning a terminal reward to individual reasoning tokens or actions remains difficult. Group-relative methods such as Group Relative Policy Optimization (GRPO) sample several attempts per prompt, but lose thei...
Wei Chen, Yu-Bing Wu, Zhou-Shi Xu et al.· 0 citations
Value-based Q-learning for continuous control can discretise the action space, but the resolution of the discretisation trades early exploration against final precision. Growing Q-Networks (GQN) grow the resolution uniformly from coarse to fine. We introduce Selective Growing Q-Networks (SGQN), which keeps decoupled Q-...
OrionMoE-Small-v1.0 is introduced, a compact tabular foundation model that combines distribution-aware representations with dataset-conditioned sparse mixture-of-experts computation that enables adaptive allocation of computation across heterogeneous tabular problems while retaining a shared computational pathway acros...
Mohamed Bouadi, Nassim Bouarour, Aditya Tanna et al.· 0 citations
Poor generalisation in deep neural networks is frequently associated with convergence to sharp minima. Sharpness-Aware Minimisation (SAM) improves generalisation of deep neural networks by minimising a combination of the training loss and the local sharpness of the loss landscape, biasing convergence towards flat minim...
Alessia Rossi, F. Campelo, Gabriel L. Oliveira· 0 citations
Momentum is a classical tool for accelerating convex optimization and a standard component of modern large language model training. Yet recent empirical work has questioned the usefulness of first-moment momentum in reinforcement learning with verifiable rewards (RLVR). We study this question through a bandit formulati...
Kaifei Wang, Zikang Shan, Liwei Wang et al.· 0 citations
We study the finite-sample convergence of Q-learning algorithms for the average-reward formulation of Reinforcement Learning (RL) with options. While options allow temporarily extended and hierarchical actions, the average-reward formulation incorporates continual tasks beyond discounted and episodic RL. We establish t...
Kintan Saha, Ahana Deb, Anders Jonsson et al.· 0 citations
