Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv:2608. 02087v2 Announce Type: replace Abstract: Post-training Large Language Models (LLMs) with Reinforcement Learning (RL) has become an important tool for improving model capabilities, but the LLM action-space structure introduces challenges distinct from classical RL, with implications for inducing exploration.
Long-horizon reasoning in recent LLMs demands that the model switch between distinct skills inside a reasoning chain, such as first doing a math derivation, then using the result to plan a schedule. We call such problems cross-skill long-horizon tasks: multi-step tasks whose steps require different reasoning skills and depend on earlier outputs.
Training LLM agents commonly relies on supervised fine-tuning from expert trajectories or online reinforcement learning over human-specified tasks with handcrafted verifiers. Though effective, both remain bottlenecked by externally specified tasks and supervision signals, limiting the scalability and diversity of agent training.
Large language model agents are commonly trained through reinforcement learning with sparse trajectory-level rewards, which offer limited guidance on how strongly individual tokens should be updated. On-Policy Self-Distillation (OPSD) addresses this by re-scoring generated tokens under a privileged replay view to obtain dense, token-level supervision.
Incentivized advertising allocates monetary or virtual rewards to drive user engagement, where a key challenge is optimizing continuous incentive magnitudes under strict global constraints. This problem is complicated by high-frequency interactions, delayed feedback, and non-Markovian user dynamics such as fatigue, which limit the effectiveness of existing uplift modeling and constrained reinforcement learning approaches.
arXiv:2602. 17976v2 Announce Type: replace-cross Abstract: In active sequential testing, also termed pure exploration, a learner is tasked with the goal to adaptively acquire information so as to identify an unknown ground-truth hypothesis with as few queries as possible.
arXiv:2608. 03527v1 Announce Type: cross Abstract: Retrieval systems help deep research agents generate high-quality answers by providing relevant documents.
arXiv:2601. 03321v3 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have substantially advanced Radiology Report Generation (RRG), yet aligning them through reinforcement learning (RL) remains challenging due to heterogeneous medical supervision.
arXiv:2608. 03483v1 Announce Type: cross Abstract: Existing chunk-based Vision-Language-Action (VLA) models execute a fixed number of actions (i.
arXiv:2608. 03031v1 Announce Type: new Abstract: Time series forecasting is fundamental to decision-making in complex systems, where future dynamics are influenced not only by historical observations but also by evolving contextual features.
arXiv:2608. 03048v1 Announce Type: cross Abstract: Long-context reasoning remains a critical bottleneck for large language models, as recent recurrent-memory approaches face two inherent challenges: sequential chunk-wise updates can overwrite early critical evidence with later irrelevant content, and serial inter-chunk dependencies limit parallelism and cause latency to increase with context length.
arXiv:2608. 03468v1 Announce Type: new Abstract: Historical tool-use trajectories provide valuable experience for large language model (LLM) agents to plan and coordinate tool usage.
arXiv:2608. 03852v1 Announce Type: new Abstract: This paper proposes FedCritic-MIMO, a communication-efficient serverless federated multi-agent reinforcement learning framework for AI-native resource control across independently deployable cell-level controllers in open and disaggregated 6G RANs.
arXiv:2608. 03316v1 Announce Type: new Abstract: On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid.
arXiv:2608. 03506v1 Announce Type: new Abstract: Self-consistency assumes the most frequent answer among sampled reasoning traces is the most reliable, but this can fail in causal reasoning: samples often repeat the same confounding error, and votes fragment across multiple valid answers, letting an invalid answer win despite a valid minority trace.
arXiv:2608. 03069v1 Announce Type: new Abstract: Deep Q-Networks (DQNs) learn value functions through bootstrapped temporal-difference updates, where future returns are approximated using a greedy maximization over next-state action values.
arXiv:2608. 03215v1 Announce Type: cross Abstract: Reinforcement learning for flow-matching text-to-speech is complicated by deterministic ODE sampling: trajectory-level policy-gradient methods typically convert the ODE into an SDE and track per-step likelihood ratios, introducing stochastic perturbations and substantial overhead.
arXiv:2608. 02993v1 Announce Type: new Abstract: (Flat) Reinforcement Learning (RL) agents face significant challenges in environments with sparse rewards that require long-horizon reasoning.
arXiv:2608. 02958v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) policies trained by behavior cloning fail silently: from the action stream alone, a collapsing rollout looks much like one making clean progress, because imitation supplies no notion of progress.
arXiv:2608. 03041v1 Announce Type: cross Abstract: Deep reinforcement learning (DRL) approaches for flexible job shop scheduling (FJSP) heavily rely on attention-centric architectures to achieve state-of-the-art performance.