Zeyi Liu

Zeyi Liu, Guangyao Liu, Yinuo Qu et al.

Human-in-the-loop reinforcement learning (HIL-RL) improves sample efficiency in real-robot manipulation through online human intervention. However, successful trajectories may include suboptimal actions that deviate from the desired task-execution path and force human intervention. Existing HIL-RL methods typically apply the consistent credit assignment principle to all transitions, uniformly propagating discounted terminal rewards through suboptimal segments, ignoring the actual contribution of each transition to task success. This overestimates Q-values for critic learning and indirectly misguides actor updates toward suboptimal behavior patterns. To this end, we propose PACT, a Preference-calibrated Actor-Critic Training framework that leverages the implicit preference signals induced by intervention to perform credit reassignment on identified suboptimal segments while directly guiding policy training for unbiased critic-actor learning. Specifically, we first design a progress model that learns from human demonstration and identifies suboptimal segments for credit correction. Then, from the human action and resampled policy action at the intervention state, we build preference pairs to define a counterfactual advantage that penalizes Bellman targets of the identified suboptimal segment, enabling directional credit calibration. Moreover, we directly align the policy with human corrective actions in the bounded mean space, providing an additional signal beyond critic-guided updates. Across five real-robot manipulation tasks, PACT improves the average success rate by 24.5% and achieves 1.3 times faster convergence, thereby improving both RL sample efficiency and performance. Code is available at https://anonymous.4open.science/r/HILRL-A1X-BC05.

Linmei Hu, Zeyi Liu, Ziwang Zhao et al.

Pre-trained Language Models (PLMs) which are trained on large text corpus via self-supervised learning method, have yielded promising performance on various tasks in Natural Language Processing (NLP). However, though PLMs with huge parameters can effectively possess rich knowledge learned from massive training text and benefit downstream tasks at the fine-tuning stage, they still have some limitations such as poor reasoning ability due to the lack of external knowledge. Research has been dedicated to incorporating knowledge into PLMs to tackle these issues. In this paper, we present a comprehensive review of Knowledge Enhanced Pre-trained Language Models (KE-PLMs) to provide a clear insight into this thriving field. We introduce appropriate taxonomies respectively for Natural Language Understanding (NLU) and Natural Language Generation (NLG) to highlight these two main tasks of NLP. For NLU, we divide the types of knowledge into four categories: linguistic knowledge, text knowledge, knowledge graph (KG), and rule knowledge. The KE-PLMs for NLG are categorized into KG-based and retrieval-based methods. Finally, we point out some promising future directions of KE-PLMs.

Zeyi Liu, Arpit Bahety, Shuran Song

The ability to detect and analyze failed executions automatically is crucial for an explainable and robust robotic system. Recently, Large Language Models (LLMs) have demonstrated strong reasoning abilities on textual inputs. To leverage the power of LLMs for robot failure explanation, we introduce REFLECT, a framework which queries LLM for failure reasoning based on a hierarchical summary of robot past experiences generated from multisensory observations. The failure explanation can further guide a language-based planner to correct the failure and complete the task. To systematically evaluate the framework, we create the RoboFail dataset with a variety of tasks and failure scenarios. We demonstrate that the LLM-based framework is able to generate informative failure explanations that assist successful correction planning.

Zeyi Liu, Zhenjia Xu, Shuran Song

We introduce BusyBoard, a toy-inspired robot learning environment that leverages a diverse set of articulated objects and inter-object functional relations to provide rich visual feedback for robot interactions. Based on this environment, we introduce a learning framework, BusyBot, which allows an agent to jointly acquire three fundamental capabilities (interaction, reasoning, and planning) in an integrated and self-supervised manner. With the rich sensory feedback provided by BusyBoard, BusyBot first learns a policy to efficiently interact with the environment; then with data collected using the policy, BusyBot reasons the inter-object functional relations through a causal discovery network; and finally by combining the learned interaction policy and relation reasoning skill, the agent is able to perform goal-conditioned manipulation tasks. We evaluate BusyBot in both simulated and real-world environments, and validate its generalizability to unseen objects and relations. Video is available at https://youtu.be/EJ98xBJZ9ek.

Ryan Punamiya, Simar Kareer, Zeyi Liu et al.

Robot learning increasingly depends on large and diverse data, yet robot data collection remains expensive and difficult to scale. Egocentric human data offer a promising alternative by capturing rich manipulation behavior across everyday environments. However, existing human datasets are often limited in scope, difficult to extend, and fragmented across institutions. We introduce EgoVerse, a collaborative platform for human data-driven robot learning that unifies data collection, processing, and access under a shared framework, enabling contributions from individual researchers, academic labs, and industry partners. The current release includes 1,362 hours (80k episodes) of human demonstrations spanning 1,965 tasks, 240 scenes, and 2,087 unique demonstrators, with standardized formats, manipulation-relevant annotations, and tooling for downstream learning. Beyond the dataset, we conduct a large-scale study of human-to-robot transfer with experiments replicated across multiple labs, tasks, and robot embodiments under shared protocols. We find that policy performance generally improves with increased human data, but that effective scaling depends on alignment between human data and robot learning objectives. Together, the dataset, platform, and study establish a foundation for reproducible progress in human data-driven robot learning. Videos and additional information can be found at https://egoverse.ai/

Songqiao Hu, Zeyi Liu, Xiao He

Concept drift detection has attracted considerable attention due to its importance in many real-world applications such as health monitoring and fault diagnosis. Conventionally, most advanced approaches will be of poor performance when the evaluation criteria of the environment has changed (i.e. concept drift), either can only detect and adapt to virtual drift. In this paper, we propose a new approach to detect real-drift in the chunk data stream with limited annotations based on concept confusion. When a new data chunk arrives, we use both real labels and pseudo labels to update the model after prediction and drift detection. In this context, the model will be confused and yields prediction difference once drift occurs. We then adopt cosine similarity to measure the difference. And an adaptive threshold method is proposed to find the abnormal value. Experiments show that our method has a low false alarm rate and false negative rate with the utilization of different classifiers.

Hongshuo Zhao, Zeyi Liu, Xiao He

Data streams in real-world industrial scenarios often contain transitional operating conditions that are uncovered during offline training, leading to significant distribution shifts. To bridge the gap between static offline models and dynamic online data, a novel asymmetric adaptation-based fault diagnosis method is proposed in this paper. Specifically, in the offline stage, we employ domain generalization techniques to extract domain-invariant features from multiple stable conditions and construct robust normalized fault prototypes as reference anchors. Subsequently, during online inference, we design an online test-time adaptation method based on a periodic prototype re-projection mechanism to dynamically update prototype positions. Furthermore, we utilize the geometric distribution derived from anchors to guide the updates of classifiers and adopt an asymmetric learning rate strategy for the feature extractor and classifier. The proposed approach ensures rapid adaptation to new transitional conditions while preserving the discriminative power inherited from the offline domain generalization initialization. Experimental results demonstrate that this mechanism effectively leverages offline generalized knowledge to guide online inference, significantly improving robustness in non-stationary environments.

Zeyi Liu, Songqiao Hu, Xiao He

Real-time safety assessment (RTSA) of dynamic systems is a critical task that has significant implications for various fields such as industrial and transportation applications, especially in non-stationary environments. However, the absence of a comprehensive review of real-time safety assessment methods in non-stationary environments impedes the progress and refinement of related methods. In this paper, a review of methods and techniques for RTSA tasks in non-stationary environments is provided. Specifically, the background and significance of RTSA approaches in non-stationary environments are firstly highlighted. We then present a problem description that covers the definition, classification, and main challenges. We review recent developments in related technologies such as online active learning, online semi-supervised learning, online transfer learning, and online anomaly detection. Finally, we discuss future outlooks and potential directions for further research. Our review aims to provide a comprehensive and up-to-date overview of real-time safety assessment methods in non-stationary environments, which can serve as a valuable resource for researchers and practitioners in this field.

Chen Li, Zeyi Liu, Limin Wang et al.

Fault diagnosis is a crucial area of research in industry. Industrial processes exhibit diverse operating conditions, where data often have non-Gaussian, multi-mode, and center-drift characteristics. Data-driven approaches are currently the main focus in the field, but continuous fault classification and parameter updates of fault classifiers pose challenges for multiple operating modes and real-time settings. Thus, a pressing issue is to achieve real-time multi-mode fault diagnosis in industrial systems. In this paper, a novel approach to achieve real-time multi-mode fault diagnosis is proposed for industrial applications, which addresses this critical research problem. Our approach uses an extended evidence reasoning (ER) algorithm to fuse information and merge outputs from different base classifiers. These base classifiers based on broad learning system (BLS) are trained to ensure maximum fault diagnosis accuracy. Furthermore, pseudo-label learning is used to update model parameters in real-time. The effectiveness of the proposed approach is demonstrated on the multi-mode Tennessee Eastman process dataset.

Zeyi Liu, Shuang Liu, Jihai Min et al.

With the rapid development of industrial intelligence and unmanned inspection, reliable perception and safety assessment for AI systems in complex and dynamic industrial sites has become a key bottleneck for deploying predictive maintenance and autonomous inspection. Most public datasets remain limited by simulated data sources, single-modality sensing, or the absence of fine-grained object-level annotations, which prevents robust scene understanding and multimodal safety reasoning for industrial foundation models. To address these limitations, InspecSafe-V1 is released as the first multimodal benchmark dataset for industrial inspection safety assessment that is collected from routine operations of real inspection robots in real-world environments. InspecSafe-V1 covers five representative industrial scenarios, including tunnels, power facilities, sintering equipment, oil and gas petrochemical plants, and coal conveyor trestles. The dataset is constructed from 41 wheeled and rail-mounted inspection robots operating at 2,239 valid inspection sites, yielding 5,013 inspection instances. For each instance, pixel-level segmentation annotations are provided for key objects in visible-spectrum images. In addition, a semantic scene description and a corresponding safety level label are provided according to practical inspection tasks. Seven synchronized sensing modalities are further included, including infrared video, audio, depth point clouds, radar point clouds, gas measurements, temperature, and humidity, to support multimodal anomaly recognition, cross-modal fusion, and comprehensive safety assessment in industrial environments.

Zeyi Liu, Songqiao Hu, Pengyu Han et al.

In recent years, online learning has attracted increasing attention due to its adaptive capability to process streaming and non-stationary data. To facilitate algorithm development and practical deployment in this area, we introduce Awesome-OL, an extensible Python toolkit tailored for online learning research. Awesome-OL integrates state-of-the-art algorithm, which provides a unified framework for reproducible comparisons, curated benchmark datasets, and multi-modal visualization. Built upon the scikit-multiflow open-source infrastructure, Awesome-OL emphasizes user-friendly interactions without compromising research flexibility or extensibility. The source code is publicly available at: https://github.com/liuzy0708/Awesome-OL.

Songqiao Hu, Zeyi Liu, Xiao He

The change in data distribution over time, also known as concept drift, poses a significant challenge to the reliability of online learning methods. Existing methods typically require model retraining or drift detection, both of which demand high computational costs and are often unsuitable for real-time applications. To address these limitations, a lightweight, fast and efficient random vector functional-link network termed Lite-RVFL is proposed, capable of adapting to concept drift without drift detection and retraining. Lite-RVFL introduces a novel objective function that assigns weights exponentially increasing to new samples, thereby emphasizing recent data and enabling timely adaptation. Theoretical analysis confirms the feasibility of this objective function for drift adaptation, and an efficient incremental update rule is derived. Experimental results on a real-world safety assessment task validate the efficiency, effectiveness in adapting to drift, and potential to capture temporal patterns of Lite-RVFL. The source code is available at https://github.com/songqiaohu/Lite-RVFL.

Longxu Dou, Qian Liu, Fan Zhou et al.

Sailor2 is a family of cutting-edge multilingual language models for South-East Asian (SEA) languages, available in 1B, 8B, and 20B sizes to suit diverse applications. Building on Qwen2.5, Sailor2 undergoes continuous pre-training on 500B tokens (400B SEA-specific and 100B replay tokens) to support 13 SEA languages while retaining proficiency in Chinese and English. Sailor2-20B model achieves a 50-50 win rate against GPT-4o across SEA languages. We also deliver a comprehensive cookbook on how to develop the multilingual model in an efficient manner, including five key aspects: data curation, pre-training, post-training, model customization and evaluation. We hope that Sailor2 model (Apache 2.0 license) will drive language development in the SEA region, and Sailor2 cookbook will inspire researchers to build more inclusive LLMs for other under-served languages.

Zixi Wang, Zeyi Liu, Nicolas Ouporov et al.

Robot-to-human object handover is an important step in many human robot collaboration tasks. A successful handover requires the robot to maintain a stable grasp on the object while making sure the human receives the object in a natural and easy-to-use manner. We propose ContactHandover, a robot to human handover system that consists of two phases: a contact-guided grasping phase and an object delivery phase. During the grasping phase, ContactHandover predicts both 6-DoF robot grasp poses and a 3D affordance map of human contact points on the object. The robot grasp poses are re-ranked by penalizing those that block human contact points, and the robot executes the highest ranking grasp. During the delivery phase, the robot end effector pose is computed by maximizing human contact points close to the human while minimizing the human arm joint torques and displacements. We evaluate our system on 27 diverse household objects and show that our system achieves better visibility and reachability of human contacts to the receiver compared to several baselines. More results can be found on https://clairezixiwang.github.io/ContactHandover.github.io

Zeyi Liu, Shuang Li, Eric Cousineau et al.

Understanding and predicting the dynamics of the physical world can enhance a robot's ability to plan and interact effectively in complex environments. While recent video generation models have shown strong potential in modeling dynamic scenes, generating videos that are both temporally coherent and geometrically consistent across camera views remains a significant challenge. To address this, we propose a 4D video generation model that enforces multi-view 3D consistency of videos by supervising the model with cross-view pointmap alignment during training. This geometric supervision enables the model to learn a shared 3D representation of the scene, allowing it to predict future video sequences from novel viewpoints based solely on the given RGB-D observations, without requiring camera poses as inputs. Compared to existing baselines, our method produces more visually stable and spatially aligned predictions across multiple simulated and real-world robotic datasets. We further show that the predicted 4D videos can be used to recover robot end-effector trajectories using an off-the-shelf 6DoF pose tracker, supporting robust robot manipulation and generalization to novel camera viewpoints.

Xiaomeng Xu, Yifan Hou, Zeyi Liu et al.

We address key challenges in Dataset Aggregation (DAgger) for real-world contact-rich manipulation: how to collect informative human correction data and how to effectively update policies with this new data. We introduce Compliant Residual DAgger (CR-DAgger), which contains two novel components: 1) a Compliant Intervention Interface that leverages compliance control, allowing humans to provide gentle, accurate delta action corrections without interrupting the ongoing robot policy execution; and 2) a Compliant Residual Policy formulation that learns from human corrections while incorporating force feedback and force control. Our system significantly enhances performance on precise contact-rich manipulation tasks using minimal correction data, improving base policy success rates by over 50\% on two challenging tasks (book flipping and belt assembly) while outperforming both retraining-from-scratch and finetuning approaches. Through extensive real-world experiments, we provide practical guidance for implementing effective DAgger in real-world robot learning tasks. Result videos are available at: https://compliant-residual-dagger.github.io/

Songqiao Hu, Zeyi Liu, Xiao He

Ensemble learning plays a crucial role in practical applications of online learning due to its enhanced classification performance and adaptable adjustment mechanisms. However, most weight allocation strategies in ensemble learning are heuristic, making it challenging to theoretically guarantee that the ensemble classifier outperforms its base classifiers. To address this issue, a performance-bounded online ensemble learning method based on multi-armed bandits, named PB-OEL, is proposed in this paper. Specifically, multi-armed bandit with expert advice is incorporated into online ensemble learning, aiming to update the weights of base classifiers and make predictions. A theoretical framework is established to bound the performance of the ensemble classifier relative to base classifiers. By setting expert advice of bandits, the bound exceeds the performance of any base classifier when the length of data stream is sufficiently large. Additionally, performance bounds for scenarios with limited annotations are also derived. Numerous experiments on benchmark datasets and a dataset of real-time safety assessment tasks are conducted. The experimental results validate the theoretical bound to a certain extent and demonstrate that the proposed method outperforms existing state-of-the-art methods.

Pengyu Han, Zeyi Liu, Shijin Chen et al.

Multi-condition fault diagnosis is prevalent in industrial systems and presents substantial challenges for conventional diagnostic approaches. The discrepancy in data distributions across different operating conditions degrades model performance when a model trained under one condition is applied to others. With the recent advancements in deep learning, transfer learning has been introduced to the fault diagnosis field as a paradigm for addressing multi-condition fault diagnosis. Among these methods, domain generalization approaches can handle complex scenarios by extracting condition-invariant fault features. Although many studies have considered fault diagnosis in specific multi-condition scenarios, the extent to which operating conditions affect fault information has been scarcely studied, which is crucial. However, the extent to which operating conditions affect fault information has been scarcely studied, which is crucial. When operating conditions have a significant impact on fault features, directly applying domain generalization methods may lead the model to learn condition-specific information, thereby reducing its overall generalization ability. This paper investigates the performance of existing end-to-end domain generalization methods under varying conditions, specifically in variable-speed and variable-load scenarios, using multiple experiments on a real-world gearbox. Additionally, a two-stage diagnostic framework is proposed, aiming to improve fault diagnosis performance under scenarios with significant operating condition impacts. By incorporating a domain-generalized encoder with a retraining strategy, the framework is able to extract condition-invariant fault features while simultaneously alleviating potential overfitting to the source domain. Several experiments on a real-world gearbox dataset are conducted to validate the effectiveness of the proposed approach.

Zeyi Liu, Cheng Chi, Eric Cousineau et al.

Audio signals provide rich information for the robot interaction and object properties through contact. This information can surprisingly ease the learning of contact-rich robot manipulation skills, especially when the visual information alone is ambiguous or incomplete. However, the usage of audio data in robot manipulation has been constrained to teleoperated demonstrations collected by either attaching a microphone to the robot or object, which significantly limits its usage in robot learning pipelines. In this work, we introduce ManiWAV: an 'ear-in-hand' data collection device to collect in-the-wild human demonstrations with synchronous audio and visual feedback, and a corresponding policy interface to learn robot manipulation policy directly from the demonstrations. We demonstrate the capabilities of our system through four contact-rich manipulation tasks that require either passively sensing the contact events and modes, or actively sensing the object surface materials and states. In addition, we show that our system can generalize to unseen in-the-wild environments by learning from diverse in-the-wild human demonstrations.

Junsheng Kong, Weizhao Li, Zeyi Liu et al.

The notion of word embedding plays a fundamental role in natural language processing (NLP). However, pre-training word embedding for very large-scale vocabulary is computationally challenging for most existing methods. In this work, we show that with merely a small fraction of contexts (Q-contexts)which are typical in the whole corpus (and their mutual information with words), one can construct high-quality word embedding with negligible errors. Mutual information between contexts and words can be encoded canonically as a sampling state, thus, Q-contexts can be fast constructed. Furthermore, we present an efficient and effective WEQ method, which is capable of extracting word embedding directly from these typical contexts. In practical scenarios, our algorithm runs 11$\sim$13 times faster than well-established methods. By comparing with well-known methods such as matrix factorization, word2vec, GloVeand fasttext, we demonstrate that our method achieves comparable performance on a variety of downstream NLP tasks, and in the meanwhile maintains run-time and resource advantages over all these baselines.