LabRobFail, a failure-centric framework for learning and evaluating robotic failure analysis in chemical laboratories, and LabRobFail-VLM, a domain-specialized vision-language model that generates structured failure diagnoses and recovery instructions, demonstrate the value of fine-grained failure understanding for closed-loop recovery and reliable laboratory autonomy.
Abstract
The deployment of embodied agents in self-driving laboratories could accelerate scientific discovery, yet their reliability is constrained by the irreversible and safety-critical nature of chemical experiments. Progress is further hindered by scarce failure data and the lack of fine-grained evaluation protocols. To address these challenges, we introduce LabRobFail, a failure-centric framework for learning and evaluating robotic failure analysis in chemical laboratories. LabRobFail-Sim injects controllable failures at the control, physics, and semantic levels, enabling the construction of LabRobFail-Data, which contains over 20,000 trajectories across 70+ task scenarios, five failure categories, and 11 fine-grained failure types. LabRobFail-Bench evaluates six capabilities spanning task understanding, failure detection, temporal localization, severity assessment, failure classification, and actionable correction. We further develop LabRobFail-VLM, a domain-specialized vision-language model that generates structured failure diagnoses and recovery instructions. On seen environments, it achieves 90.83% failure-detection accuracy and 77.21% temporal-localization accuracy, substantially outperforming general-purpose VLMs. When integrated as a real-time supervisor, it improves downstream task success rates by 4-16 percentage points, demonstrating the value of fine-grained failure understanding for closed-loop recovery and reliable laboratory autonomy. Our code and data are available at https://github.com/Su-ISE-2001/SciRobo
Imitation learning has shown increasing promise for autonomous robotic surgery, yet safe deployment remains challenging due to the safety-critical nature of surgical tasks and the complexity and variability of surgical environments. Failure detection is therefore an essential safeguard, but its development remains difficult due to the challenges of scarce failure data, highly variable manipulation dynamics, and the need to balance missed detections against disruptive false alarms. To address these challenges, we introduce FoMo-FD (Flow-Matching World Model for Failure Detection), a failure detection method that learns nominal short-horizon visual dynamics with an action-conditioned flow-matching world model. FoMo-FD scores the inverse-transport nonconformity of observed endpoint latents, enabling window-level detection of visual-action inconsistencies without requiring failure demonstrations. Detection thresholds are obtained by conformal calibration on successful executions, yielding task-specific alarms without assuming future failure types. We evaluate FoMo-FD on four surgically relevant manipulation tasks with twenty failure modes across simulation and real-world experiments using the da Vinci Research Kit (dVRK). Results show that FoMo-FD outperforms observation-level anomaly baselines and a prediction-error variant of the same world model, with the wrist-camera view achieving the strongest performance, including a 96.6% failure detection rate (FDR) at a 1.3% false alarm rate (FAR).
Zhefeng Huang, Yilin Cai, Ankit B. Patel et al.· 2 citations
Foundation-model policies for robotic manipulation are advancing rapidly on task success, but rigorous evaluation of whether they succeed safely is still lacking. We introduce ManiGuard, a specification-grounded framework for evaluating and improving the safety of foundation-model manipulation, comprising the ManiGuard-Bench task suite and a paired safety-annotated trajectory-generation pipeline. ManiGuard-Bench organizes six contact-rich household task families into 200 locked base tasks along a skill $\times$ constraint taxonomy, with safety specified independently of task success. Each task is evaluated under one in-distribution and four single-axis out-of-distribution perturbations that hold the safety specification fixed, giving 1,000 locked scenarios. Every rollout is runtime-checked by LTL$_f$-grounded automaton monitors over physics-grounded predicates rather than learned classifiers or LLM judges, in simulation and on a physical Franka platform. The pipeline pairs an automated motion-planning generator with human teleoperation, annotated by the same per-step monitor, and directly supports safety-aware fine-tuning; we release 8,000 safety-annotated demonstrations, 40 per base task. Benchmarking zero-shot and fine-tuned VLAs across more than 23,000 rollouts, we find: (i) safety must be evaluated independently of task success, as 6-21% of successful rollouts violate the specification; (ii) fine-tuning on our suite raises safe task completion from near zero to 7.5-29.8% and engaged-and-safe behavior from 16-40% to 51-72%; but (iii) a gap remains that scaling demonstrations does not close, with 21-42% of engaged rollouts still violating, two of six families below 2% safe success for every policy, and these failures persisting under distribution shift and on hardware.
We introduce LabEvolver, a training-free framework that equips safe and grounded wet-lab agents with episodic memory from execution experience. LabEvolver couples a state-grounded inner trial loop for adaptive perception, online planning, and safety validation with an outer evolution loop that distills completed trajectories into reusable skill, strategy, and safety experience. On robotic solution-preparation tasks, LabEvolver demonstrates real-world feasibility, reducing pH-regulation completion time and safety-gate intercepts by 48.2% and 60.0%, respectively. On ALFWorld, it further improves cumulative success rate within 20 steps from 76.2% with ReAct to 91.4% over 500 continual tasks, showing generality beyond wet-lab settings. These results support learn-by-doing experience evolution as a feasible path toward closed-loop automated scientific discovery. The project page is available at https://andygao6186.github.io/LabEvolver/.
Jingya Wang, Yuyang Gao, Liuzhenghao Lv et al.· 0 citations
Vision–Language–Action (VLA) policies are increasingly evaluated on language-conditioned robotic manipulation benchmarks, but success rate alone often obscures runtime-interface alignment, the repeatability of observation-degradation effects, and the recovery-relevant semantics of failures. This study proposes ROEP (Robotics-Oriented Evaluation Protocol), a VLA-targeted deployment-oriented evaluation protocol that converts rollout outcomes into claim-level evidence for closed-loop robotic manipulation. ROEP first verifies clean-condition evaluability and runtime fidelity of the sensor-to-action execution interface, then evaluates controlled visual perturbations, repeated-run reference variability, timeout-dominant failures, and recovery/shield support. We apply ROEP to OpenVLA, X-VLA, and VLA-Adapter on eight LIBERO Goal and Object tasks, producing a 24-row evaluation matrix with complete clean and medium-perturbation evidence. Under the evaluated LIBERO simulation setting, the results show that runtime-interface fidelity and checkpoint provenance are important for interpreting X-VLA and OpenVLA Object results, while VLA-Adapter maintains strong Goal-suite performance but exhibits a substantial Object-suite clean-to-perturbation drop dominated by timeout termination. ROEP therefore clarifies which deployment-relevant claims are supported, withheld, or insufficiently evidenced by the available rollouts without certifying open-world deployment safety or recovery success. The protocol complements existing VLA benchmarks by reporting success rate together with runtime fidelity, repeatability, failure semantics, and recovery/shield support evidence.
Sangwoo Han, Hyunguk Choi· Italian National Conference...· 0 citations
Vision-Language Models (VLMs) are increasingly used to evaluate robot manipulation outcomes, but existing benchmarks offer limited evidence of cross-domain generalization. We introduce FailBench, a benchmark for robot failure detection comprising 2,197 manipulation attempts across 14 public sources (12 real-world, 2 simulated). In FailBench, 75% of failures occur naturally, and six real-world sources come from non-failure-detection datasets. Evaluating 13 VLM-based detectors, we find the best model achieves only 0.77 mean balanced accuracy. Notably, models fine-tuned for failure detection consistently underperform general-purpose VLMs and their own pretrained baselines. Performance depends heavily on required visual evidence: models approach saturation when outcomes depend on observable object motion, but degrade to near-chance (<0.60 balanced accuracy) on contact-intensive assembly tasks. Error analysis reveals a systematic bias toward predicting success under ambiguous evidence, which persists even with increased reasoning effort. Finally, we show that input-level intervention--spatially localizing and cropping outcome-relevant regions--improves the top detector by 2.4 percentage points without extra training.
The deployment of autonomous robotic systems in chemistry laboratories is accelerating experimental workflows and providing the foundational data for AI-driven scientific discovery. However, despite the success of data-driven methods in acquiring dexterous skills, safety remains a primary barrier to their deployment in high-risk domains, such as early-stage materials chemistry experiments. Specifically, learning-based policies frequently struggle to distinguish between safe and unsafe actions, leading to overconfident extrapolation and potentially catastrophic failures. To mitigate these safety risks, we propose SAFE-CHEM, an uncertainty-aware framework designed for robust, learning-based robotic chemists. Our approach leverages an ensemble of recurrent neural network-based imitation learning policies to quantify epistemic uncertainty online through the variance of action predictions. By characterising the success-conditioned density of this variance using kernel density estimation, we introduce a hybrid control architecture that autonomously switches from the learned policy to a deterministic, rule-based backup controller when uncertainty exceeds a calibrated safety threshold. We evaluate SAFE-CHEM across three fundamental laboratory manipulation tasks, where our empirical results demonstrate that this hybrid strategy improves overall task success rates and reduces critical safety violations compared to traditional single-policy baselines. Finally, we demonstrate the practical viability of the framework through zero-shot sim-to-real transfer onto a physical Franka Production 3 robot manipulator.
Laura Jones, Shazil Shahzad, Ayesha Sana et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
