SPAARS: Safer RL Policy Alignment through Abstract Exploration and Refined Exploitation of Action Space

• URL: http://arxiv.org/abs/2603.09378v2
• Date: Wed, 11 Mar 2026 05:45:21 GMT
• Title: SPAARS: Safer RL Policy Alignment through Abstract Exploration and Refined Exploitation of Action Space
• Abstract summary: offline-to-online reinforcement learning (RL) offers a promising paradigm for robotics by pre-training policies on safe, offline demonstrations and fine-tuning them via online interaction.<n>We introduce SPAARS, a curriculum learning framework that initially constrains exploration to the low-dimensional latent manifold for sample-efficient, safe behavioral improvement.<n>We prove an upper bound on the exploitation gap using the Performance Difference Lemma, establish that latent-space policy gradients achieve provable variance reduction over raw-space exploration, and show that concurrent behavioral cloning during the latent phase directly controls curriculum transition stability.
• Score: 1.8763872698583384
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Offline-to-online reinforcement learning (RL) offers a promising paradigm for robotics by pre-training policies on safe, offline demonstrations and fine-tuning them via online interaction. However, a fundamental challenge remains: how to safely explore online without deviating from the behavioral support of the offline data? While recent methods leverage conditional variational autoencoders (CVAEs) to bound exploration within a latent space, they inherently suffer from an exploitation gap -- a performance ceiling imposed by the decoder's reconstruction loss. We introduce SPAARS, a curriculum learning framework that initially constrains exploration to the low-dimensional latent manifold for sample-efficient, safe behavioral improvement, then seamlessly transfers control to the raw action space, bypassing the decoder bottleneck. SPAARS has two instantiations: the CVAE-based variant requires only unordered (s,a) pairs and no trajectory segmentation; SPAARS-SUPE pairs SPAARS with OPAL temporal skill pretraining for stronger exploration structure at the cost of requiring trajectory chunks. We prove an upper bound on the exploitation gap using the Performance Difference Lemma, establish that latent-space policy gradients achieve provable variance reduction over raw-space exploration, and show that concurrent behavioral cloning during the latent phase directly controls curriculum transition stability. Empirically, SPAARS-SUPE achieves 0.825 normalized return on kitchen-mixed-v0 versus 0.75 for SUPE, with 5x better sample efficiency; standalone SPAARS achieves 92.7 and 102.9 normalized return on hopper-medium-v2 and walker2d-medium-v2 respectively, surpassing IQL baselines of 66.3 and 78.3 respectively, confirming the utility of the unordered-pair CVAE instantiation.

Related papers

• The Geometry of Refusal: Linear Instability in Safety-Aligned LLMs [1.4323566945483497]
  We introduce Contrastive Logit Steering ( CLS), a zero-optimization framework that isolates the "refusal direction"<n> CLS operates directly on the output distribution, serving as a diagnostic probe for alignment.<n>Our experiments on 7 model families reveal that safety implementation is architecturally deterministic.
  arXiv  Detail & Related papers  (2026-06-21T22:04:48Z)
• StepOPSD: Step-Aware Online Preference Distillation for Agent Reinforcement Learning [6.365332042924078]
  We present StepOPSD, a post-rollout preference self-distillation framework that takes the agent step as the unit of credit redistribution.<n>StepOPSD decomposes trajectories into action-centered step segments, rescoring them under hindsight-enriched teacher contexts.<n>Results reveal a consistent two-knob law: smaller _clip acts as a broadly stabilizing local trust region, whereas the optimal global mixing strength _mix remains task-dependent.
  arXiv  Detail & Related papers  (2026-05-26T15:07:03Z)
• High-Dimensional Latents Should Be Diagnosed Through Phase Structure [56.362776482614976]
  We study autoencoder and variational-autoencoder latent spaces through the lens of spin-glass theory.<n>We formalize a latent-space spin-glass dictionary for a fixed decoder.<n>We show that driving the latent system toward the edge-of-stability of the topological trivialization regime has concrete downstream consequences.
  arXiv  Detail & Related papers  (2026-05-23T06:17:23Z)
• Improving Search Agent with One Line of Code [68.58667107354253]
  Tool-based Agentic Reinforcement Learning (TARL) has emerged as a promising paradigm for training search agents to interact with external tools.<n>We propose textbfSearch textbfAgent textbfPolicy textbfOptimization (textbfSAPO), which stabilizes training via a conditional token-level KL constraint.
  arXiv  Detail & Related papers  (2026-03-10T04:07:39Z)
• RetroAgent: From Solving to Evolving via Retrospective Dual Intrinsic Feedback [54.39884046754265]
  Large language model (LLM)-based agents trained with reinforcement learning (RL) have shown strong potential on complex interactive tasks.<n>We introduce RetroAgent, an online RL framework that empowers agents to master complex interactive environments not just by solving, but by evolving.
  arXiv  Detail & Related papers  (2026-03-09T16:23:33Z)
• Latent Policy Steering through One-Step Flow Policies [34.06099184809882]
  offline reinforcement learning (RL) allows robots to learn from offline datasets without risky exploration.<n>Latent Policy Steering (LPS) enables high-fidelity latent policy improvement by backpropagating original-action-space Q-gradients through a differentiable one-step MeanFlow policy.<n>Across OGBench and real-world robotic tasks, LPS achieves state-of-the-art performance and consistently outperforms behavioral cloning and strong latent steering baselines.
  arXiv  Detail & Related papers  (2026-03-05T15:38:08Z)
• Hybrid Belief Reinforcement Learning for Efficient Coordinated Spatial Exploration [3.0222726254970174]
  Pure model-based approaches provide structured uncertainty estimates but lack adaptive policy learning.<n>This paper presents a hybrid belief-reinforcement learning framework to address this gap.<n>Results show 10.8% higher cumulative reward and 38% faster convergence over baselines.
  arXiv  Detail & Related papers  (2026-03-04T00:00:34Z)
• Controllable Exploration in Hybrid-Policy RLVR for Multi-Modal Reasoning [88.42566960813438]
  CalibRL is a hybrid-policy RLVR framework that supports controllable exploration with expert guidance.<n>CalibRL increases policy entropy in a guided manner and clarifies the target distribution.<n>Experiments across eight benchmarks, including both in-domain and out-of-domain settings, demonstrate consistent improvements.
  arXiv  Detail & Related papers  (2026-02-22T07:23:36Z)
• Function-Space Decoupled Diffusion for Forward and Inverse Modeling in Carbon Capture and Storage [65.51149575007149]
  We present Fun-DDPS, a generative framework that combines function-space diffusion models with differentiable neural operator surrogates for both forward and inverse modeling.<n>Fun-DDPS produces physically consistent realizations free from the high-frequency artifacts observed in joint-state baselines.
  arXiv  Detail & Related papers  (2026-02-12T18:58:12Z)
• On GRPO Collapse in Search-R1: The Lazy Likelihood-Displacement Death Spiral [59.14787085809595]
  We identify Lazy Likelihood Displacement (LLD) as the core mechanism driving this failure.<n>LLD emerges early and triggers a self-reinforcing LLD Death Spiral, where declining likelihood leads to low-confidence responses.<n>We propose a lightweight likelihood-preserving regularization LLDS for GRPO that activates only when a trajectory's likelihood decreases.
  arXiv  Detail & Related papers  (2025-12-03T19:41:15Z)
• Human-in-the-loop Online Rejection Sampling for Robotic Manipulation [55.99788088622936]
  Hi-ORS stabilizes value estimation by filtering out negatively rewarded samples during online fine-tuning.<n>Hi-ORS fine-tunes a pi-base policy to master contact-rich manipulation in just 1.5 hours of real-world training.
  arXiv  Detail & Related papers  (2025-10-30T11:53:08Z)
• Guardian: Decoupling Exploration from Safety in Reinforcement Learning [12.966077380225856]
  Hybrid offline--online reinforcement learning (O2O RL) promises both sample efficiency and robust exploration, but suffers from instability due to distribution shift between offline and online data.<n>We introduce RLPD-GX, a framework that decouples policy optimization from safety enforcement.<n>We empirically show state-of-the-art performance on Atari-100k, achieving a normalized mean score of 3.02 (+45% over prior hybrid methods) with stronger safety and stability.
  arXiv  Detail & Related papers  (2025-10-26T22:25:47Z)
• Bridging the Synthetic-Real Gap: Supervised Domain Adaptation for Robust Spacecraft 6-DoF Pose Estimation [13.83897333268682]
  Spacecraft Pose Estimation is a fundamental capability for autonomous space operations such as rendezvous, docking, and in-orbit docking.<n>Existing domain adaptation approaches aim to mitigate this issue but often underperform when a modest number of labeled target samples are available.<n>We propose the first Supervised Domain Adaptation (SDA) framework tailored for SPE keypoint regression.
  arXiv  Detail & Related papers  (2025-09-17T08:03:05Z)
• Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding [55.107555305760954]
  We propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. Our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.
  arXiv  Detail & Related papers  (2024-01-17T09:01:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.