reinforcement learning

Training LLMs with Reinforcement Learning over Digital Twin Representations for Reasoning-Intensive Surgical VideoQA

Human-In-The-Loop Multi-agent Ventilator Decision Support with Contextual Bandit Preference Learning

SutureFormer: Learning Surgical Trajectories via Goal-Conditioned Offline RL in Pixel Space

Supply chain resilience prediction and industrial economic shock transmission path analysis integrating knowledge graphs and deep reinforcement learning

Reinforcement Learning-Driven Dynamic Trading Strategies for Financial Markets

This paper presents the RL-DynTrade framework by using a cutting-edge deep reinforcement learning method, Proximal Policy Optimization (PPO), with a Deep Q Network (DQN) agent to dynamically adapt to changing risk-reward dynamics. PPO enables real-time, fine-grained, risk-reward adaptation via an actor-critic design wi...

Reading Journey to the West as an ML Development Pipeline

This paper presents a continuous, stage-by-stage structural correspondence between the 100-chapter classical Chinese novel Journey to the West (Xiyouji, c. 1592) and the modern development pipeline for Large Language Models (LLMs). Without asserting that the historical author predicted modern technology, the analysis d...

Retrospective registration of a systematic review protocol: AI-Based Intrusion Detection and Prevention: A Systematic Review and Adversarial-Validity Audit of Benchmark Datasets

Retrospective registration of a systematic review protocol. This review was conducted and submitted prior to registration; no prospective registration was performed. The review synthesizes AI-based techniques for intrusion detection and prevention, drawing on 71 peer-reviewed studies published between 2019 and 2025, se...

Reinforcement Learning-Based Intelligent Adaptive Protection Relays: A Systematic Comparative Review of Algorithms, Applications, and Performance Metrics

The accelerating integration of inverter-based resources (IBR) and distributed generation (DG)—with IBR penetration now exceeding 40% in many distribution networks and fault-current contribution curtailed to 1.0–1.5 p.u. versus 5–10 p.u. for synchronous generators—has measurably reduced the efficacy of conventional pow...

From pixels to actions: grounding in vision-language models

Vision-Language Models (VLMs) have demonstrated impressive capabilities across a broad range of perception tasks, yet systematic weaknesses remain concerning fine-grained spatial reasoning, precise visual localisation, or grounded action in interactive environments. This thesis investigates the architectural and traini...

Classical Reinforcement Learning Algorithms

UniRG-CXR: model weights and inference code

Archival snapshot of the UniRG-CXR chest X-ray report generation model (weights and inference code), exactly as used in the study 'Scaling medical imaging report generation with multimodal reinforcement learning'. Mirrors the Hugging Face repository microsoft/UniRG-CXR at git revision cd5a95fa837c92180e14ed0ddd7f079150...

Reinforcement Learning for Reasoning Models