Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2506. 10630v3 Announce Type: replace-cross Abstract: To advance time series forecasting (TSF), various methods have been proposed to improve prediction accuracy, evolving from statistical techniques to data-driven deep learning architectures.
arXiv:2606. 04305v1 Announce Type: new Abstract: We study online learning with an additional offline dataset in the stochastic linear bandit setting.
arXiv:2605. 04356v2 Announce Type: replace-cross Abstract: Reinforcement learning with verifiable rewards has been used to elicit impressive performance from language models in many domains.
arXiv:2605. 20654v2 Announce Type: replace-cross Abstract: While Large Language Models (LLMs) demonstrate remarkable capabilities, they remain susceptible to sophisticated, multi-step jailbreak attacks that circumvent conventional surface-level safety alignment by exploiting the internal generation process.
arXiv:2606. 04749v1 Announce Type: cross Abstract: Safe robot control requires maximizing return while satisfying safety constraints.
arXiv:2603. 21180v4 Announce Type: replace Abstract: Sequential experimental design under expensive, gradient-free objectives is a central challenge in computational statistics: evaluation budgets are tightly constrained and information must be extracted efficiently from each observation.
arXiv:2606. 04120v1 Announce Type: cross Abstract: Conversational agents that serve as lifelong companions must maintain persistent memory across all interactions.
arXiv:2606. 04574v1 Announce Type: new Abstract: This study aims to determine whether the application of Deep Reinforcement Learning (DRL) as a specialized execution overlay can enhance pair trading in highly volatile cryptocurrency markets.
arXiv:2209. 15448v3 Announce Type: replace Abstract: As AI becomes more prevalent throughout society, effective methods of integrating humans and AI systems that leverage their respective strengths and mitigate risk have become an important priority.
arXiv:2606. 04492v1 Announce Type: new Abstract: Cooperative Multi-Agent Reinforcement Learning (MARL) frequently suffers from severe reward sparsity and exploration bottlenecks.
arXiv:2606. 04272v1 Announce Type: new Abstract: The standard LLM training pipeline applies reinforcement learning (RL) only after pre-training and supervised fine-tuning (SFT).
arXiv:2510. 27191v5 Announce Type: replace-cross Abstract: Planning under partial observability is an essential capability of autonomous robots.
arXiv:2606. 04632v1 Announce Type: new Abstract: Mechanical ventilation for Acute Respiratory Distress Syndrome (ARDS) requires balancing competing physiological goals, including oxygenation, lung protection, and acid-base homeostasis.
Large language models (LLMs) are rigorously aligned to refuse harmful requests, a process that inherently cultivates a latent capacity to evaluate and recognize unsafe content. In this work, we reveal that this advanced safety awareness inadvertently introduces a fatal vulnerability.
Scaling reinforcement learning (RL) to diverse multitask settings remains a central challenge. While recent advances in model-based RL achieve strong performance, they rely on planning and complex training pipelines, making it unclear which components are essential for scalability.
Large language models are increasingly evaluated by other models, raising a natural question: can a model predict how a judge will score its own output? We find that the ability is largely present before any targeted training: prompted few-shot, a base model already predicts an external judge's multi-attribute quality scores on open-ended responses well above chance across three benchmarks.
Reinforcement learning with verifiable rewards (e. g.
Mitigating social bias in Large Language Models (LLMs) presents a distinct alignment challenge: unlike verifiable tasks, bias lacks a single ground truth, creating a high-variance, subjective reward landscape. Previous preference-based fine-tuning methods have major trade-offs: Direct Preference Optimization (DPO) is limited by the lack of exploration inherent in offline training, while Proximal Policy Optimization (PPO) can lead to training instability due to potentially unreliable critic estimates.
Agent Skills today consist largely of free-form prose requiring the agent to read, interpret, and re-derive how to act in every session. This imposes two compounding costs: reduced reliability on implementation-heavy tasks, and difficulty in skill creation and improvement, since editing prose is a fragile process that both humans and agents struggle with, particularly for domain-specific procedural knowledge underrepresented in model training.
Temporal credit assignment is central to both biological and artificial intelligence, yet its interaction with non-linear function approximation is poorly understood. We identify a systematic failure mode in deep reinforcement learning (RL) termed Trace-Mediated Peak Bias (TMPB).