arXiv Machine Learning
Aug 12
Boundary-Seeking Policy Gradient for Safe Reinforcement Learning
arXiv:2608. 10204v1 Announce Type: new Abstract: Safe reinforcement learning maximizes reward subject to safety constraints.
Reinforcement learning
Policy optimisation, reward modelling and RLHF — how models are trained by feedback rather than by labels.
arXiv Machine Learning
Aug 12
Partially Observable Learning for Multi-Platform Dispatch Optimization
arXiv:2608. 10897v1 Announce Type: new Abstract: Instant delivery platforms have become a critical component of urban logistics, increasingly relying on crowdsourced couriers to fulfill highly dynamic orders.
arXiv AI
Aug 12
VDC-Agent: When Video Detailed Captioners Evolve Themselves via Agentic Self-Reflection
arXiv:2511. 19436v2 Announce Type: replace-cross Abstract: Existing Video Detailed Captioning (VDC) methods predominantly rely on costly human annotations or distillation from powerful proprietary models, creating a dependency on external supervision.
arXiv Machine Learning
Aug 12
Efficient Hypergradient Descent for Inverse Reinforcement Learning
arXiv:2608. 11052v1 Announce Type: new Abstract: Inverse reinforcement learning (IRL) aims to recover a reward function under which the resulting policy reproduces the behavior observed in expert demonstrations.
arXiv AI
Aug 12
Coordinating the Unknown Lipschitz Constant in Multiplayer Bandits
arXiv:2608. 10526v1 Announce Type: cross Abstract: Motivated by decentralized applications, we study cooperative multi-agent bandits in continuous (Lipschitz) action spaces when the Lipschitz constant is unknown.
arXiv AI
Aug 12
Access Timing as Scaffolding: A Reinforcement Learning Approach to GenAI in Education
arXiv:2605. 15850v3 Announce Type: replace-cross Abstract: In recent years, generative AI (GenAI) in educational settings has become ubiquitous in university students' daily lives, despite its potential to induce over-reliance, metacognitive disengagement, and diminished learning when used unrestrictedly.
arXiv Machine Learning
Aug 12
Generalized Linear Markov Decision Process
arXiv:2506. 00818v2 Announce Type: replace-cross Abstract: Offline reinforcement learning for longitudinal studies often faces two linked challenges: rewards may be binary or bounded, and reward observations may be available only for a subset of trajectories or time points even when the corresponding state-action-next-state histories are available.
arXiv AI
Aug 12
Procedural Fairness Failures in RLHF from Preference Averaging
arXiv:2608. 10126v1 Announce Type: cross Abstract: Reinforcement Learning from Human Feedback (RLHF) aggregates heterogeneous preferences into a single reward model, assuming preference homogeneity.
arXiv AI
Aug 12
Reference-Free Post-Training of Open Large Language Models for Multilingual Machine Translation
arXiv:2608. 10812v1 Announce Type: cross Abstract: We study reference-free post-training for multilingual machine translation with open large language models.
Hugging Face Trending Papers
Aug 12
Reinforcing Step-level Reasoning for Effective Self-Correction in LLMs
Achieving effective self-correction, where models verify and correct their own mistakes, remains a fundamental challenge for large language models (LLMs). In this work, we propose Self-Fix Step-DPO (SFS-DPO), a reinforcement learning based, two-stage framework for step-level self-verification and self-correction.
Hugging Face Trending Papers
Aug 12
When Offline Evaluation Misleads: A Diagnostic Protocol for Reward and Policy Selection in Delayed-Feedback Contextual Bandits
Personalizing marketing messages with contextual multi-armed bandits (CMABs) drives real business value, yet the objective that ultimately matters - a downstream conversion - is observed only weeks later, too late to drive online learning. Teams therefore train the bandit on a fast proxy reward, and separately must judge whether a contextual bandit is worth its complexity over sending one best message.
Hugging Face Trending Papers
Aug 11
Scheduling Mixed RL Rollouts Beyond Prefix Locality
Modern reinforcement learning (RL) post-training pipelines for large language models (LLMs) increasingly combine rollout workloads across multiple domains and feedback paradigms. Prefix-aware routing improves inference efficiency through cache reuse and load balancing, but it does not control how heterogeneous rollout sessions compete for KV-cache capacity.
Hugging Face Trending Papers
Aug 11
SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models
Agent skills represent a standardized format for packaging procedural knowledge and domain expertise, serving within agent harness systems as an essential mechanism to continually constrain a language model's behavior space for repeatable, high-quality task execution. However, because strong closed-source models entail high inference costs, current popular agent harnesses, such as Codex and OpenClaw, remain prohibitively expensive when deploying these skills to accomplish real-world tasks.
Hugging Face Trending Papers
Aug 11
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones. We propose SafeCap, a reinforcement-learning framework that aligns LVLMs through learned self-captioning.
Hugging Face Trending Papers
Aug 11
Exploration-Driven Personalized Federated Reinforcement Learning via Intrinsic Motivation
Personalized Federated Reinforcement Learning (PFRL) takes a decentralized approach to storing and accessing information based on past experiences while keeping each client's data private during the learning of each client's policy. Many current methods for PFRL rely heavily on exploiting existing reinforcement learning reward signals to derive an optimal policy for each client, thereby neglecting exploration in non-stationary or sparse-reward environments.
arXiv Machine Learning
Aug 11
Learning from Environmental Feedback: Credit Assignment across Multiple Timescales for Agentic Reinforcement Learning
arXiv:2608. 08255v1 Announce Type: new Abstract: Agentic reinforcement learning (RL) often suffers from delayed and sparse rewards in real-world environments.
arXiv AI
Aug 11
RISE-RL: Rubric-Informed Selective Exploration for Open-Ended Reinforcement Learning
arXiv:2608. 09123v1 Announce Type: new Abstract: Aligning Large Language Models (LLMs) for open-ended tasks is challenging because responses must satisfy multidimensional criteria without following a single correct generation trajectory.
arXiv Machine Learning
Aug 11
DreOPD: Degraded-Reference Extrapolative On-Policy Distillation for Flow-matching Models
arXiv:2608. 09233v1 Announce Type: new Abstract: Flow-matching models are now a mainstream method to image generation, but its adaptation to diverse downstream scenarios typically relies on post-training, which may cause conflicts among task-specific optimization objectives.
arXiv AI
Aug 11
CoRE: Consensus Rewards via Equilibrium for Test-Time Reinforcement Learning
arXiv:2608. 09324v1 Announce Type: new Abstract: On unlabeled test data, reinforcement learning lacks a ground-truth reward; test-time RL methods derive one from the model's own roll-outs, rewarding those that match the majority vote over $N$ sampled answers.
arXiv Machine Learning
Aug 11
Bayesian Symbolic Regression with Entropic Reinforcement Learning
arXiv:2608. 09617v1 Announce Type: new Abstract: Symbolic regression is the problem of finding an algebraic expression describing a stochastic dependence of a target variable on a set of inputs.