LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning

By Bo Liu, Yifeng Zhu, Chongkai Gao, Yihao Feng, Qian Liu, Yuke Zhu, Peter Stone. Neural Information Processing Systems, 2023

1,206 citations recorded by the source corpus.

Lifelong learning offers a promising paradigm of building a generalist agent that learns and adapts over its lifespan. Unlike traditional lifelong learning problems in image and text domains, which primarily involve the transfer of declarative knowledge of entities and concepts, lifelong learning in decision-making (LLDM) also necessitates the transfer of procedural knowledge, such as actions and behaviors. To advance research in LLDM, we introduce LIBERO, a novel benchmark of lifelong learning for robot manipulation. Specifically, LIBERO highlights five key research topics in LLDM: 1) how to efficiently transfer declarative knowledge, procedural knowledge, or the mixture of both; 2) how to design effective policy architectures and 3) effective algorithms for LLDM; 4) the robustness of a lifelong learner with respect to task ordering; and 5) the effect of model pretraining for LLDM. We develop an extendible procedural generation pipeline that can in principle generate infinitely many tasks. For benchmarking purpose, we create four task suites (130 tasks in total) that we use to investigate the above-mentioned research topics. To support sample-efficient learning, we provide high-quality human-teleoperated demonstration data for all tasks. Our extensive experiments present several insightful or even unexpected discoveries: sequential finetuning outperforms existing lifelong learning methods in forward transfer, no single visual encoder architecture excels at all types of knowledge transfer, and naive supervised pretraining can hinder agents' performance in the subsequent LLDM. Check the website at https://libero-project.github.io for the code and the datasets.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/98cfd7b1b29453c4e82536f5afdc6ddc58bbb1b3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success (2025, neutral): “The continuous actions variant with L1 regression shows negligible difference in efficiency since the additional MLP action head adds minimal computational [25].”
CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (2025, neutral): “We conduct evaluations across three complementary settings: the LIBERO benchmark [34] for evaluation in simulation environments, the Bridge-V2 platform [54] with its dataset of 45k robot demonstrations, and the Franka-Tabletop setup with a stationary, table-mounted Franka Emika Panda robot with…”
RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots (2024, background): “Some are limited to tabletop manipulation [33, 15, 6, 28], but RoboCasa, along with others [21, 40, 26, 27, 10, 34], support mobile manipulation.”
Online Continual Learning For Interactive Instruction Following Agents (2024, neutral): “…strategies ( e.g. , reinforcement learning (Khetarpal et al., 2018; Wołczyk et al., 2021; Xie & Finn, 2022), imitation learning (Mendez et al., 2018; Gao et al., 2021), etc. ) and task formulations ( e.g. , manipulation (Yang et al., 2022; Liu et al., 2023), walking (Zhou et al., 2022b), etc. ).”
SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Model (2025, neutral): “We present the evaluation of SpatialVLA on the LIBERO simulation benchmark [36], which consists of a set of diverse robotic manipulation tasks in simulated environments.”
RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation (2025, neutral): “…for multi-task and meta-reinforcement learning, CALVIN pairs long-horizon, language-conditioned instructions with rich sensor suites [27], LIBERO [25] defines 130 lifelong-learning tasks with high-quality human teleoperation data, and RoboVerse [13] unifies multiple simulators and embodiments…”
UniVLA: Learning to Act Anywhere with Task-centric Latent Actions (2025, neutral): “The LIBERO benchmark [48] comprises four task suites specifically designed to facilitate research on life-long learning in robotic manipulation.”
SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics (2025, positive): “SmolVLA outperforms other VLA-based approaches such as Octo (Team et al., 2024) and OpenVLA (Kim et al., 2024), as well as the diffusion policy baseline across both LIBERO and Meta-World.”