4arXiv cs.AI (Artificial Intelligence)·
Adaptive safety margins for visual robot navigation via context-conditioned safety critic
A new arXiv paper proposes a context-conditioned safety critic that learns adaptive clearance preferences for ranking diffusion-based trajectory proposals in indoor robot navigation. The critic decomposes into safety, efficiency, and distance-constraint matching terms, trained with privileged ESDF geometry in simulation and distilled into a perception-only selector via teacher-student learning. The method achieves top success rate and SPL on PointGoal navigation benchmarks (HM3D, MP3D) and transfers zero-shot to a Unitree G1 humanoid robot without task-specific tuning.
Related guides (1)
Related events (8)
Verifiable Belief-Space Neural Safety Filters for Interactive Robotics via Conformal Prediction
This paper proposes an algorithmic framework to certify high-probability safety of belief-space safety filters (BeliefSF) in interactive robotics, addressing the challenge that neural approximations and runtime inference errors make formal guarantees difficult. The approach uses conformal prediction focused on regions where inference is reliable, preserving standard sample complexity while certifying a less conservative filter. Evaluation on a simulated human-vehicle interaction benchmark demonstrates the method produces significantly more permissive safety guarantees than a standard conformal prediction baseline.
CARLA-GS: Modular pipeline for photorealistic autonomous driving corner-case synthesis using LLMs and Gaussian splatting
CARLA-GS is a new modular framework for synthesizing safety-critical corner cases for autonomous driving evaluation, decoupling visual representation (editable Gaussian scenes), semantic reasoning (multi-agent LLM for waypoint generation), and physics simulation (CARLA with PID control). The pipeline reconstructs real driving scenes from the Waymo Open Dataset, uses LLMs to identify risky interactions and generate intent-level trajectories, then re-projects simulated states into Gaussian scenes for photorealistic ego-centric rendering. The approach addresses a key gap in AV safety evaluation: generating spatiotemporally consistent, physically feasible, and semantically meaningful rare scenarios at scale.
EvolveNav: Self-evolving memory and preflection for zero-shot object-goal navigation
EvolveNav is a new framework for Zero-Shot Object-Goal Navigation (ZS-OGN) that enables test-time improvement through a self-evolving agentic rule memory built from past trajectories. A retrieval strategy based on upper confidence bound balances semantic relevance and historical success when selecting rules, while a memory-guided preflection module forecasts action outcomes before execution to reduce inefficient exploration. The method achieves a 10.1% improvement in success rate over existing zero-shot baselines with fewer unnecessary steps.
SafeCtrl-RL: Inference-Time Adaptive Behaviour Control for LLMs via RL-Driven Prompt Optimisation
SafeCtrl-RL is a framework for controlling LLM safety at inference time without retraining or modifying model parameters. It formulates dialogue generation as a sequential decision process where an RL agent dynamically selects prompt adjustment strategies based on contextual feedback, iteratively suppressing unsafe outputs. The authors frame this as 'inference-time behavioural unlearning' and report improvements in safety and response quality across multiple LLMs and unsafe dialogue scenarios, outperforming existing prompt-based optimisation baselines.
VLESA: Vision-Language Embodied Safety Agent for Real-Time Human Activity Monitoring
Researchers introduce VLESA, a framework that monitors human activities from egocentric video and triggers real-time safety interventions when dangerous actions are predicted. The system addresses intent-dependent safety — where identical actions can be safe or dangerous depending on context — using a goal-conditioned safety Q-filter trained via GRPO and an intent-action prediction agent. On the ASIMOV-2.0 benchmark, VLESA achieves higher intervention accuracy than baselines, with the Q-filter improving action safety by over 41 percentage points through goal-conditioned constrained decoding.
IA-VQC-DPC: Intervention-aware quantum predictive control with safety attribution for learned policies
A new arXiv preprint introduces Intervention-Aware Variational Quantum Differentiable Predictive Control (IA-VQC-DPC), a framework that trains variational quantum circuit policies under a primal-dual intervention budget to penalize over-reliance on downstream safety filters (Control-Barrier-Function projections). The work also proposes a safety-attribution protocol that decomposes trajectory corrections into policy-level versus filter-level contributions, enabling measurement of whether a policy has genuinely learned safe behavior or is merely being silently repaired by its safety layer. Experiments on BOPTEST building-control emulators show the quantum policy achieves significantly lower pre-filter violations than a matched classical policy at equal parameter budget, with a notable negative result: a learned energy head is only safe when paired with a distribution-aware runtime guard.
Safety Gym: OpenAI Releases RL Safety Constraint Benchmark Suite
OpenAI released Safety Gym, a suite of environments and tools designed to measure progress in training reinforcement learning agents that respect safety constraints during training. The toolkit targets the challenge of constrained RL, where agents must optimize objectives without violating specified safety boundaries. This represents an early formal effort by OpenAI to provide standardized benchmarking infrastructure for safe RL research.
ContextRL: Context-aware reinforcement learning improves grounding in agentic and multimodal LLMs
Researchers introduce ContextRL, a reinforcement learning method that trains LLMs to select the context that supports a given query-answer pair from two highly similar candidates, rather than supervising only final answers. The approach constructs contrastive context pairs in two domains: coding agent trajectories (1k pairs) and multimodal image pairs (7k pairs). ContextRL achieves +2.2% average gains over standard GRPO on 5 long-horizon benchmarks and +1.8% across 12 visual QA benchmarks, with ablations showing the gains stem from the context-selection objective rather than the contrastive data alone.