5arXiv cs.CL (Computation and Language)·
RCT dataset enables contact-sequence-aware evaluation of tactile generalization for robot manipulation
Researchers introduce RCT (Robotic Contact Tactile), an open-source dataset of 29,279 tactile frames collected from robot presses on 122 industrial materials, paired with vision and language annotations via three DIGIT sensors. The paper exposes a critical data leakage problem in existing tactile benchmarks: frame-random splits allow near-duplicate observations across train/test, inflating performance by up to 17.7 Recall@1 points. Proper held-out-material evaluation reveals sharp performance drops (Recall@1 of 25.1%), framing novel-material generalization as a central unsolved challenge for robotic tactile perception. The dataset and evaluation protocol are publicly released.
Related guides (1)
Related events (8)
LeRobot Community Datasets: The "ImageNet" of Robotics — When and How?
Hugging Face's LeRobot blog post discusses the vision and current state of building a large-scale community robotics dataset analogous to ImageNet for computer vision. The post examines what it would take to create a standardized, scalable dataset repository for robot learning, drawing on the LeRobot ecosystem. It addresses data collection formats, community contribution workflows, and the open challenges in making such a resource practically useful for training generalizable robot policies.
AutoDex: Automated real-world dexterous grasping data collection with 4.8x throughput over teleoperation
AutoDex is an automated system for collecting physically-labeled dexterous grasping data at scale, closing the perception-execution-labeling-reset loop without human intervention. The system uses 20-camera dense perception to handle hand-object occlusion, executes collision-monitored motions, and actively resets objects between trials. Across 100 objects and two robot hand platforms, AutoDex achieves 4.8x throughput versus teleoperation and yields 76% grasp success from its validated database versus 34% for simulation-only validation. Code and data will be publicly released.
TREAD: VLM-based re-labelling framework improves robot policy generalization via dataset augmentation
TREAD (Task Robustness via Re-Labelling Vision-Action Robot Data) is a scalable framework that uses pretrained Vision-Language Models to augment existing robotics datasets without new data collection. The approach decomposes demonstrations into sub-tasks, segments videos accordingly, and generates linguistically diverse instruction labels, enriching language-action pair diversity. Evaluations on the LIBERO benchmark show improved generalization to novel tasks and goals, addressing a key limitation of current robot learning policies.
OCR-Robust benchmark evaluates VLM robustness to visual perturbations on OCR-reasoning tasks
Researchers introduce OCR-Robust, a benchmark of 812 samples designed to evaluate how vision-language models handle OCR-reasoning tasks under controlled visual degradation. The benchmark covers documents, scene text, charts, geometry, and tables, applying 5 perturbation types at 3 severity levels each, and evaluates 18 models using metrics including Relative Corruption Retention and a composite Corruption Robustness Index. Key findings show that higher clean accuracy does not guarantee robustness, and that chart and table inputs are substantially more fragile under perturbation than document-like inputs.
E-TTS: Embodied test-time scaling framework for robotic manipulation with history-aware iterative refinement
Researchers introduce E-TTS, a modular test-time scaling framework for robotic manipulation that unifies reasoning and action scaling via history-aware iterative refinement and vision-language verifiers. The framework addresses two gaps in prior work: underexplored reasoning scaling mechanisms and inadequate use of historical context in long-horizon sequential tasks. Evaluated across 4 benchmarks, 6 environments, 3 embodiments, and 4 base vision-language-action models, E-TTS achieves up to 33.14% improvement in simulation and 26.62% in real-world scenarios without additional expert data or retraining.
Solving Rubik's Cube with a Robot Hand via Reinforcement Learning and Automatic Domain Randomization
OpenAI trained neural networks to solve a Rubik's Cube using a dexterous robot hand, with training conducted entirely in simulation via reinforcement learning. A new technique called Automatic Domain Randomization (ADR) enables the system to generalize to real-world physical perturbations not seen during training. The work demonstrates that sim-to-real transfer can achieve unprecedented dexterity in manipulation tasks.
FACTR 2: Neural External Torque Estimation enables force-aware robot policy learning without dedicated sensors
Researchers introduce NEXT (Neural External Torque Estimation), a data-driven method that estimates external joint torques on commodity robot arms without dedicated force sensors, training in one minute from ten minutes of free-motion data. Combined with Force-Informed Re-Sampling Training (FIRST), which up-samples contact segments during behavior cloning, the system outperforms prior force-aware policies by over 17% in task progress across five long-horizon manipulation tasks. The work lowers the hardware barrier for contact-rich robot manipulation by bringing force-feedback teleoperation and policy learning to off-the-shelf arms.
LeRobot Goes to Driving School: World's Largest Open-Source Self-Driving Dataset
Hugging Face's LeRobot framework has been extended to include what is claimed to be the world's largest open-source self-driving dataset, released via a blog post on March 11, 2025. The dataset is intended to accelerate research in autonomous driving by providing large-scale, openly accessible driving data. This represents a significant expansion of LeRobot beyond its original robotics manipulation focus into the autonomous vehicle domain.