Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2508. 06659v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) agents often struggle to generalize to new tasks and contexts without updating their parameters, mainly because their learned representations and policies are overfit to the specifics of their training environments.
arXiv:2511. 07046v4 Announce Type: replace-cross Abstract: Deploying continuous-control reinforcement learning policies on embedded hardware requires meeting tight latency and power budgets.
arXiv:2606. 08543v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) improves large language model reasoning but often suffers from rapid policy-entropy collapse, where the policy prematurely concentrates on narrow high-probability reasoning paths.
arXiv:2511. 20397v2 Announce Type: replace Abstract: We present BLINQ, a new model-based algorithm that learns the Whittle indices of an indexable, communicating and unichain Markov Decision Process (MDP).
arXiv:2505. 21457v2 Announce Type: replace-cross Abstract: Active vision, also known as active perception, refers to actively selecting where and how to look in order to gather task-relevant information.
arXiv:2601. 23286v4 Announce Type: replace-cross Abstract: While recent video diffusion models (VDMs) produce visually impressive results, they fundamentally struggle to maintain 3D structural consistency, often resulting in object deformation or spatial drift.
arXiv:2602. 02572v2 Announce Type: replace-cross Abstract: Existing alignment methods directly use the reward model learned from user preference data to optimize an LLM policy, subject to KL regularization with respect to the base policy.
arXiv:2603. 10823v2 Announce Type: replace-cross Abstract: Deep generative models can help with data scarcity and privacy by producing synthetic training data, but they struggle in low-data, imbalanced tabular settings to fully learn the complex data distribution.
arXiv:2605. 22781v2 Announce Type: replace-cross Abstract: LLM-powered AI agents require high-frequency state exploration (e.
arXiv:2606. 07988v1 Announce Type: new Abstract: Large language models (LLMs) increasingly rely on reward models to align their outputs with diverse user preferences.
arXiv:2605. 03862v4 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards has become a common way to improve explicit reasoning in large language models, but final-answer correctness alone does not reveal whether the reasoning trace is faithful, reliable, or useful to the model that consumes it.
arXiv:2605. 26078v3 Announce Type: replace Abstract: Wasserstein policy gradient (WPG) is a policy optimization method for reinforcement learning (RL) that exploits the optimal-transport geometry of action distributions.
arXiv:2509. 07779v2 Announce Type: replace-cross Abstract: We study decentralized online Riemannian optimization over manifolds with possibly positive curvature, going beyond the Hadamard manifold setting.
arXiv:2510. 13554v2 Announce Type: replace-cross Abstract: The reasoning pattern of Large language models (LLMs) remains opaque, and reinforcement learning (RL) typically applies uniform credit across an entire generation, blurring the distinction between pivotal and routine steps.
arXiv:2606. 09073v1 Announce Type: cross Abstract: Reinforcement learning from human feedback (RLHF) is bottlenecked by \emph{reward hacking}, where the policy exploits errors in a proxy reward model (RM) and produces high RM scores without genuine quality gains.
arXiv:2602. 01357v2 Announce Type: replace Abstract: Self-play post-training methods has emerged as an effective approach for finetuning large language models and turn the weak language model into strong language model without preference data.
arXiv:2606. 08379v1 Announce Type: new Abstract: This study addresses the optimal execution of large stock sell programs by introducing TT-DAC-PS (Twin-Target Deterministic Actor-Critic with Policy Smoothing), a deterministic actor-critic architecture that combines twin exponential-moving-average critic targets with pessimistic min backup, TD3-style target policy smoothing noise, delayed actor updates, and conservative Q regularisation to curb overestimation.
arXiv:2603. 14762v3 Announce Type: replace-cross Abstract: We study supervisory switching control for partially-observed linear dynamical systems.
arXiv:2605. 00327v2 Announce Type: replace-cross Abstract: In large language model (LLM)-based recommendation systems, direct preference optimization (DPO) effectively aligns recommendations with user preferences, requiring multi-negative objective functions to leverage abundant implicit-feedback negatives and sharpen preference boundaries.
arXiv:2606. 08450v1 Announce Type: new Abstract: Financial portfolio trading is naturally formulated as a reinforcement learning problem, where an agent sequentially rebalances assets under changing market conditions to balance return, risk, and transaction costs.