Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2610.28977v1 Announce Type: cross Abstract: In-context reinforcement learning (ICRL) exploits the impressive in-context learning ability of large language models to zero-shot generalize to new,...
arXiv:2610.29149v1 Announce Type: cross Abstract: Learning safe policies that generalize to unseen environments is difficult under bounded actuation, where an unsafe decision may become unrecoverable...
arXiv:2610.29187v1 Announce Type: cross Abstract: Motivated by ten years of Learning-Based Control Invited Sessions at the IEEE Conference on Decision and Control (CDC), this tutorial paper reviews k...
arXiv:2610.30105v1 Announce Type: cross Abstract: Sequential decision-making under partial observability is difficult because agents must explore to infer hidden, task-relevant information before act...
arXiv:2610.30132v1 Announce Type: cross Abstract: Reinforcement learning (RL) is an effective way to strengthen the long-context reasoning of large language models (LLMs), but it needs training data...
arXiv:2604.18936v2 Announce Type: replace Abstract: Despite the growing application of large language models (LLMs) to theoretical physics, there has been little academic exploration of how domain-sp...
arXiv:2607.24900v2 Announce Type: replace Abstract: Language model alignment aims to make model behavior reflect desirable properties such as helpfulness, safety, and instruction following. Existing...
arXiv:2604.06192v2 Announce Type: replace-cross Abstract: Recent work uses entropy-based signals at multiple representation levels to study reasoning in large language models, but the field remains l...
arXiv:2605.13252v2 Announce Type: replace-cross Abstract: We study multiple change point localization under bandit feedback. An unknown piecewise-constant function on a compact interval can be querie...
arXiv:2610.29189v1 Announce Type: cross Abstract: Large language models (LLMs) demonstrate impressive emergent reasoning capabilities, yet they remain constrained by fixed model parameters and finite...
arXiv:2610.22671v2 Announce Type: replace Abstract: As assistants move from reactive answering toward proactive intervention, memory changes role: it no longer only supplies the answer to a query, it...
arXiv:2610.05195v3 Announce Type: replace Abstract: The next generation of multimodal research agents must reason over long-lived research histories rather than short model completions. During a sing...
arXiv:2602.06232v2 Announce Type: replace Abstract: Game balancing is a longstanding challenge requiring repeated playtesting, expert intuition, and extensive manual tuning. We introduce RuleSmith, a...
arXiv:2603.17685v4 Announce Type: replace Abstract: Balancing policy expressiveness with the exploration-exploitation trade-off is a core challenge in online Reinforcement Learning (RL). While Stocha...
arXiv:2605.01513v2 Announce Type: replace Abstract: Designing therapeutic messenger RNA (mRNA) requires generating full-length transcripts that optimally balance intracellular stability and translati...
arXiv:2601.17454v3 Announce Type: replace-cross Abstract: Tabular studies are a common way to compare independent and centralized multi-agent learners, because they avoid function approximation. A na...
arXiv:2510.06198v4 Announce Type: replace Abstract: Despite the remarkable reasoning capabilities of large language models, they still struggle with one-shot relation extraction without predefined re...
arXiv:2610.28829v1 Announce Type: new Abstract: Instruction-guided video generative models offer a scalable solution for stress-testing autonomous driving systems by simulating diverse environmental...
arXiv:2610.29552v1 Announce Type: new Abstract: Fine-grained multimodal intelligence requires models to both understand a designated region and localize it from language. Existing methods learn these...
arXiv:2610.29590v1 Announce Type: new Abstract: Cinematic multi-shot video generation requires both narrative planning and iterative refinement. Existing methods either lack narrative reasoning or re...