Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2608.24485v1 Announce Type: cross Abstract: Automated parking commonly assumes marked slots and short approach maneuvers. Delivery and service vehicles, however, may need to reach an operator-s...
arXiv:2505.01361v3 Announce Type: replace Abstract: Temporal difference (TD) learning is a foundational algorithm in reinforcement learning (RL). For nearly forty years, TD learning has served as a w...
The paper introduces SMITH, a reinforcement learning framework that jointly trains large language models to create and use tools within a single policy. By alternating between build and use tasks and employing separate reward signals for schema, code, and outcome failures, SMITH enables a 4B Qwen3 model to achieve state‑of‑the‑art accuracy on procedural reasoning benchmarks, outperforming larger untrained models and improving performance on downstream tasks when its tools are applied.
arXiv:2608.24859v1 Announce Type: new Abstract: Multi-task vehicle routing problem (VRP) solvers seek to handle multiple VRP variants within a single unified model, avoiding the need to train a separ...
arXiv:2412.02520v4 Announce Type: replace-cross Abstract: Connected automated vehicles (CAVs) equipped with adaptive cruise control (ACC) create new opportunities for highway congestion mitigation. T...
arXiv:2601.19969v2 Announce Type: replace-cross Abstract: Human-in-the-loop guidance has emerged as an effective approach for accelerating online reinforcement learning (RL) in real-world manipulatio...
The paper introduces AnTrap, a benchmark that injects dynamic perturbations into Android GUI agent execution to evaluate robustness against runtime anomalies. It presents a taxonomy of anomalies across four layers—State, Thinking, Action, and Round—with ten subcategories, and a pipeline that maintains task solvability while adding realistic adversarial conditions. Experiments on 16 leading GUI models show universal vulnerability, and reinforcement learning can mitigate some traps but not deep contextual ones like state deadlock.
arXiv:2504.19342v4 Announce Type: replace-cross Abstract: Reinforcement Learning from Human Feedback (RLHF) has become a pivotal paradigm in artificial intelligence to align large models with human p...
arXiv:2511.02605v3 Announce Type: replace Abstract: Shielding is widely used to enforce safety in reinforcement learning (RL), ensuring that an agent's actions remain compliant with formal specificat...
MetaRAG introduces a belief-action aligned policy optimization framework for agentic retrieval-augmented generation (RAG). It incorporates Verify-first Action Generation and Internal Belief Probing to assess whether the current evidence is sufficient before taking an action, and uses a consistency reward gated by answer correctness to guide training. Experiments on seven public QA benchmarks demonstrate that MetaRAG improves the accuracy-efficiency trade-off over existing RL-based agentic RAG baselines, with benefits that transfer across research settings, optimizers, and model backbones.
The paper introduces AdaptRubric, a Coarse-to-Fine Rubrics Framework designed to create task‑adaptive judging criteria for GUI reward modeling. It first retrieves a category‑level coarse rubric by mapping instructions to a GUI task family, then generates an instance‑level fine rubric that captures specific values, scopes, and constraints from the instruction. Experiments show that AdaptRubric outperforms existing reward agents, improving F1 by 3.6 points and achieving a 4.23‑point task‑success gain under a matched image budget.
arXiv:2604.06621v2 Announce Type: replace-cross Abstract: Dr. David Blackwell was a mathematician and statistician of the first rank, whose contributions to statistical theory, game theory, and decis...
Marginalized importance weighting evaluates a target policy by reweighting offline state-action samples with its discounted occupancy ratio, characterized by an adjoint Bellman equation. Existing mini...
Outcome-supervised search agents learn when and how to retrieve evidence, but terminal rewards neither localize intermediate errors nor redirect an ongoing trajectory before those errors compound. Tre...
Large language model (LLM) agents are trained with reinforcement learning (RL) for complex decision-making tasks. However, most RL-trained agents remain episodic and cannot accumulate reusable knowled...
Reinforcement learning can align diffusion models with human preferences and task-specific objectives, but endpoint rewards do not specify how an intermediate denoising prediction should change. We in...
Large Language Model (LLM) agents increasingly solve long-horizon tasks through multi-turn interactions with users and external tools. In these settings, relevant task information often unfolds over t...
Massively parallel simulation changes the data regime in which off-policy reinforcement learning (RL) is trained, challenging stabilizers designed for data-limited replay. Through controlled experimen...
Search-augmented reasoning remains difficult for small language models. On-policy distillation (OPD) from trained teachers offers a promising direction, but suffers from two issues: (1) high-quality m...
Safeguarding language model agents requires assessing complete execution trajectories under context-dependent safety policies. Existing policy-aware safeguards mainly rely on prompting or supervised f...