Searcharxiv⌕ Search

arXiv subjects

Tianfu Li

Publications and source records attributed to Tianfu Li.

14 recordsLinked to original sources

Ego-CEVR: Learning Cross-Embodiment Visual Representations from Human Egocentric Videos

Egocentric videos offer a low-cost, scalable source of manipulation experience for robot learning. However, existing methods often fail to tailor their use of egocentric videos and robot demonstrations to the differences in their formats and information content. We present Ego-CEVR, a data-native framework for hierarchical visual representation learning that assigns distinct learning objectives to global and interaction representations. The global module learns task context through future video prediction, while the interaction module uses this context to interpret wrist observations and learns local execution information through robot action supervision. The distinct learning objectives enable independent pretraining on large-scale human and robot egocentric videos to learn task context, and training on real-robot demonstrations to align the learned representations with fine-grained robot actions. Experiments show leading performance on LIBERO-Plus and RoboTwin 2.0, while real-robot evaluations demonstrate generalization across tasks and embodiments.

cs.RO↗

Elastic-WAM: Adaptive Spatiotemporal Imagination for Efficient Robot Control

World-Action Models improve robot control by predicting future visual dynamics, but iterative video denoising incurs substantial inference overhead. Recent approaches reduce this cost through partial future denoising, yet typically rely on fixed refinement budgets. Our empirical observations reveal that preferred refinement depths vary across tasks and states, while different future regions exhibit varying relevance to action generation. Motivated by these findings, we propose Elastic-WAM, which adaptively allocates world refinement computation across temporal denoising time and spatial future space. A unified spatiotemporal refinement demand field governs when to stop world refinement, how far to advance each denoising step, and which regions to refresh. We parameterize this field with a shared temporal basis and spatial amplitudes, and learn it through a three-stage action-grounded supervised procedure. At inference, the learned field guides recurrent sparse world refinement to improve task success while reducing inference overhead. Evaluations across diverse simulation benchmarks and real-world long-horizon manipulation demonstrate state-of-the-art control performance with reduced inference latency.

cs.RO↗

PMTRM: Pseudo-Memory Temporal Re-encoding Module for Embodied Policy Learning

Robotic manipulation often contains repeated motions whose local observations look similar at different phases. When these phases require different actions, a policy that relies mainly on the current observation may repeat completed motions or switch phases at the wrong time. To address this phase ambiguity, we present the Pseudo-Memory Temporal Re-encoding Module (PMTRM), a lightweight plug-in module with only 7.61M parameters that encodes a bounded history of executed states and actions into a latent sequence for existing policies. To help distinguish phases, a temporal heterogeneity objective penalizes positive similarity between distant positions in this sequence, while anchor and reconstruction losses preserve information needed for action prediction. The reconstruction decoder is used only during training, leaving the temporal re-encoder to supply history to the policy at inference. We train the module progressively on synthetic sequences and robot data, then jointly with the policy, using temporal masking to accommodate partial histories. This integration retains the original action head and action space and adds auxiliary losses to the original policy loss. Experiments with multiple policy backbones in simulation and on a real robot show improved task success on tasks with phase ambiguity, with little additional computation.

cs.RO↗

MVG-WAM: Multiple View Geometry-Aware World-Action Modeling for Robotic Manipulation

World-Action Models (WAMs) couple visual dynamics with action prediction, bringing the rich priors of pretrained video models to robotic manipulation. However, their multi-view interfaces typically tile images or concatenate tokens, leaving the geometric relationships among synchronized cameras implicit. This makes it harder to connect global scene context with the local geometry required for interaction. We introduce the Multi-View Geometry-Aware World-Action Model (MVG-WAM), which organizes these observations as related projections of one physical world rather than separate images on a canvas. Our model combines an epipolar-constrained global state with view-indexed geometric states jointly inferred from synchronized observations. Camera-aware routing supplies each video region with its corresponding geometric context and the shared global state, explicitly structuring the representation used for action prediction. We further ground the geometry-aware representation in metric scale through multi-horizon future-depth supervision, without requiring depth decoding during action rollout. MVG-WAM achieves average success rates of 99.1% on LIBERO and 92.07% on RoboTwin 2.0, demonstrating competitive performance across both benchmarks. Real-world experiments on Cobot Magic further demonstrate a 91.3% success rate across 150 trials spanning three manipulation tasks.

cs.RO↗

SLIP-VLA: Single-Step Latent Imagination for Policy Learning in Vision-Language-Action Models

Vision-Language-Action models are increasingly effective for robotic manipulation, yet most predict actions directly from current observations without explicitly modeling future scene evolution. Recent methods introduce future prediction to improve action generation, but dense future modeling often requires expensive iterative denoising, while one-step alternatives can underperform their multi-step counterparts. To reconcile efficient future modeling with strong action performance, we present SLIP-VLA, a policy learning framework that equips VLA models with a Single-Step Latent Imagination for future-aware action prediction. SLIP-VLA obtains temporally dense future latent representations with a single denoising update, and we improve the perceptual sufficiency of these representations by aligning intermediate latents with future geometric and semantic features. We further improve their control sufficiency through action-conditioned latent world modeling and inverse dynamics modeling, explicitly coupling latent transitions with robot actions. SLIP-VLA achieves state-of-the-art performance across diverse simulation benchmarks and real-world manipulation tasks, while its single-step latent imagination takes only 12 ms.

cs.RO↗

IRPO: Boosting Image Restoration via Post-training GRPO

Post-training has become effective for high-level generation, but its role in low-level vision remains underexplored. Existing image restoration methods often rely on fixed pixel-wise fitting to ground-truth images, which can lead to over-smoothing and weak generalization. We propose IRPO, a GRPO-based post-training framework for deterministic restoration models. IRPO is built around two axes: data formulation and reward modeling. For data formulation, we select the 30% underperforming samples from the pre-training stage, which improves both accuracy and training efficiency. For reward modeling, we combine fidelity-oriented and quality-aware feedback with three components: a General Reward for structural fidelity, an Expert Reward that uses a Vision-Language Model as a coarse visual-quality judge, and a Restoration Reward for task-specific low-level cues. Experiments on six in-domain and five out-of-domain (OOD) benchmarks show that IRPO improves the AdaIR baseline by 0.93 dB on in-domain tasks and 3.43 dB on OOD settings. Our code can be shown in https://github.com/HaoxuanXU1024/IRPO.

cs.CV↗

HCSG: Human-Centric Semantic-Geometric Reasoning for Vision-Language Navigation

VLN has achieved remarkable progress by scaling data and model capacity. However, the assumption of a static environment breaks down in real-world indoor scenarios, where robots inevitably encounter dynamic pedestrians. Existing human-aware approaches typically treat humans merely as moving obstacles based on implicit visual cues, lacking the explicit reasoning required to interpret human intentions or maintain social norms. To address this, we propose HCSG, the first human-centric framework for VLN. This framework provides a robust foundation for safe, socially intelligent navigation in dynamic human-robot environments that shifts the paradigm from passive collision avoidance to active human behavior understanding. Specifically, HCSG introduces a unified Human Understanding Module that synergizes two key capabilities: (i) geometric forecasting, which predicts human pose and trajectory to anticipate future motion dynamics; and (ii) semantic interpretation, which leverages a Vision-Language Model (VLM) to generate natural language descriptions of human actions and intentions. These semantic-geometric representations are fused into the agent's topological map for instruction-conditioned planning. Furthermore, a social distance loss is introduced to enforce socially compliant interaction distances. Extensive experiments on the HA-VLNCE benchmark demonstrate that HCSG significantly outperforms state-of-the-art methods, achieving a 14% improvement in Success Rate and a 34% reduction in Collision Rate. Our project can be seen at https://haoxuanxu1024.github.io/HCSG/.

cs.RO↗

P$^{3}$Nav: End-to-End Perception, Prediction and Planning for Vision-and-Language Navigation

In Vision-and-Language Navigation (VLN), an agent is required to plan a path to the target specified by the language instruction, using its visual observations. Consequently, prevailing VLN methods primarily focus on building powerful planners through visual-textual alignment. However, these approaches often bypass the imperative of comprehensive scene understanding prior to planning, leaving the agent with insufficient perception or prediction capabilities. Thus, we propose P$^{3}$Nav, a novel end-to-end framework integrating perception, prediction, and planning in a unified pipeline to strengthen the VLN agent's scene understanding and boost navigation success. Specifically, P$^{3}$Nav augments perception by extracting complementary cues from object-level and map-level perspectives. Subsequently, our P$^{3}$Nav predicts waypoints to model the agent's potential future states, endowing the agent with intrinsic awareness of candidate positions during navigation. Conditioned on these future waypoints, P$^{3}$Nav further forecasts semantic map cues, enabling proactive planning and reducing the strict reliance on purely historical context. Integrating these perceptual and predictive cues, a holistic planning module finally carries out the VLN tasks. Extensive experiments demonstrate that our P$^{3}$Nav achieves new state-of-the-art performance on the REVERIE, R2R-CE, and RxR-CE benchmarks.

cs.RO↗

Enhancing Vision-Language Navigation with Multimodal Event Knowledge from Real-World Indoor Tour Videos

Vision-Language Navigation (VLN) agents often struggle with long-horizon reasoning in unseen environments, particularly when facing ambiguous, coarse-grained instructions. While recent advances use knowledge graph to enhance reasoning, the potential of multimodal event knowledge inspired by human episodic memory remains underexplored. In this work, we propose an event-centric knowledge enhancement strategy for automated process knowledge mining and feature fusion to solve coarse-grained instruction and long-horizon reasoning in VLN task. First, we construct YE-KG, the first large-scale multimodal spatiotemporal knowledge graph, with over 86k nodes and 83k edges, derived from real-world indoor videos. By leveraging multimodal large language models (i.e., LLaVa, GPT4), we extract unstructured video streams into structured semantic-action-effect events to serve as explicit episodic memory. Second, we introduce STE-VLN, which integrates the above graph into VLN models via a Coarse-to-Fine Hierarchical Retrieval mechanism. This allows agents to retrieve causal event sequences and dynamically fuse them with egocentric visual observations. Experiments on REVERIE, R2R, and R2R-CE benchmarks demonstrate the efficiency of our event-centric strategy, outperforming state-of-the-art approaches across diverse action spaces. Our data and code are available on the project website https://sites.google.com/view/y-event-kg/.

cs.RO↗

A Novel Unsupervised Graph Wavelet Autoencoder for Mechanical System Fault Detection

Reliable fault detection is an essential requirement for safe and efficient operation of complex mechanical systems in various industrial applications. Despite the abundance of existing approaches and the maturity of the fault detection research field, the interdependencies between condition monitoring data have often been overlooked. Recently, graph neural networks have been proposed as a solution for learning the interdependencies among data, and the graph autoencoder (GAE) architecture, similar to standard autoencoders, has gained widespread use in fault detection. However, both the GAE and the graph variational autoencoder (GVAE) have fixed receptive fields, limiting their ability to extract multiscale features and model performance. To overcome these limitations, we propose two graph neural network models: the graph wavelet autoencoder (GWAE), and the graph wavelet variational autoencoder (GWVAE). GWAE consists mainly of the spectral graph wavelet convolutional (SGWConv) encoder and a feature decoder, while GWVAE is the variational form of GWAE. The developed SGWConv is built upon the spectral graph wavelet transform which can realize multiscale feature extraction by decomposing the graph signal into one scaling function coefficient and several spectral graph wavelet coefficients. To achieve unsupervised mechanical system fault detection, we transform the collected system signals into PathGraph by considering the neighboring relationships of each data sample. Fault detection is then achieved by evaluating the reconstruction errors of normal and abnormal samples. We carried out experiments on two condition monitoring datasets collected from fuel control systems and one acoustic monitoring dataset from a valve. The results show that the proposed methods improve the performance by around 3%~4% compared to the comparison methods.

eess.SP↗

Smart filter aided domain adversarial neural network for fault diagnosis in noisy industrial scenarios

The application of unsupervised domain adaptation (UDA)-based fault diagnosis methods has shown significant efficacy in industrial settings, facilitating the transfer of operational experience and fault signatures between different operating conditions, different units of a fleet or between simulated and real data. However, in real industrial scenarios, unknown levels and types of noise can amplify the difficulty of domain alignment, thus severely affecting the diagnostic performance of deep learning models. To address this issue, we propose an UDA method called Smart Filter-Aided Domain Adversarial Neural Network (SFDANN) for fault diagnosis in noisy industrial scenarios. The proposed methodology comprises two steps. In the first step, we develop a smart filter that dynamically enforces similarity between the source and target domain data in the time-frequency domain. This is achieved by combining a learnable wavelet packet transform network (LWPT) and a traditional wavelet packet transform module. In the second step, we input the data reconstructed by the smart filter into a domain adversarial neural network (DANN). To learn domain-invariant and discriminative features, the learnable modules of SFDANN are trained in a unified manner with three objectives: time-frequency feature proximity, domain alignment, and fault classification. We validate the effectiveness of the proposed SFDANN method based on two fault diagnosis cases: one involving fault diagnosis of bearings in noisy environments and another involving fault diagnosis of slab tracks in a train-track-bridge coupling vibration system, where the transfer task involves transferring from numerical simulations to field measurements. Results show that compared to other representative state of the art UDA methods, SFDANN exhibits superior performance and remarkable stability.

eess.SP↗

Filter-informed Spectral Graph Wavelet Networks for Multiscale Feature Extraction and Intelligent Fault Diagnosis

Intelligent fault diagnosis has been increasingly improved with the evolution of deep learning (DL) approaches. Recently, the emerging graph neural networks (GNNs) have also been introduced in the field of fault diagnosis with the goal to make better use of the inductive bias of the interdependencies between the different sensor measurements. However, there are some limitations with these GNN-based fault diagnosis methods. First, they lack the ability to realize multiscale feature extraction due to the fixed receptive field of GNNs. Secondly, they eventually encounter the over-smoothing problem with increase of model depth. Lastly, the extracted features of these GNNs are hard to understand owing to the black-box nature of GNNs. To address these issues, a filter-informed spectral graph wavelet network (SGWN) is proposed in this paper. In SGWN, the spectral graph wavelet convolutional (SGWConv) layer is established upon the spectral graph wavelet transform, which can decompose a graph signal into scaling function coefficients and spectral graph wavelet coefficients. With the help of SGWConv, SGWN is able to prevent the over-smoothing problem caused by long-range low-pass filtering, by simultaneously extracting low-pass and band-pass features. Furthermore, to speed up the computation of SGWN, the scaling kernel function and graph wavelet kernel function in SGWConv are approximated by the Chebyshev polynomials. The effectiveness of the proposed SGWN is evaluated on the collected solenoid valve dataset and aero-engine intershaft bearing dataset. The experimental results show that SGWN can outperform the comparative methods in both diagnostic accuracy and the ability to prevent over-smoothing. Moreover, its extracted features are also interpretable with domain knowledge.

eess.SP↗

Deep Learning Algorithms for Rotating Machinery Intelligent Diagnosis: An Open Source Benchmark Study

With the development of deep learning (DL) techniques, rotating machinery intelligent diagnosis has gone through tremendous progress with verified success and the classification accuracies of many DL-based intelligent diagnosis algorithms are tending to 100\%. However, different datasets, configurations, and hyper-parameters are often recommended to be used in performance verification for different types of models, and few open source codes are made public for evaluation and comparisons. Therefore, unfair comparisons and ineffective improvement may exist in rotating machinery intelligent diagnosis, which limits the advancement of this field. To address these issues, we perform an extensive evaluation of four kinds of models, including multi-layer perception (MLP), auto-encoder (AE), convolutional neural network (CNN), and recurrent neural network (RNN), with various datasets to provide a benchmark study within the same framework. We first gather most of the publicly available datasets and give the complete benchmark study of DL-based intelligent algorithms under two data split strategies, five input formats, three normalization methods, and four augmentation methods. Second, we integrate the whole evaluation codes into a code library and release this code library to the public for better development of this field. Third, we use specific-designed cases to point out the existing issues, including class imbalance, generalization ability, interpretability, few-shot learning, and model selection. By these works, we release a unified code framework for comparing and testing models fairly and quickly, emphasize the importance of open source codes, provide the baseline accuracy (a lower bound) to avoid useless improvement, and discuss potential future directions in this field. The code library is available at https://github.com/ZhaoZhibin/DL-based-Intelligent-Diagnosis-Benchmark.

eess.SP↗

WaveletKernelNet: An Interpretable Deep Neural Network for Industrial Intelligent Diagnosis

Convolutional neural network (CNN), with ability of feature learning and nonlinear mapping, has demonstrated its effectiveness in prognostics and health management (PHM). However, explanation on the physical meaning of a CNN architecture has rarely been studied. In this paper, a novel wavelet driven deep neural network termed as WaveletKernelNet (WKN) is presented, where a continuous wavelet convolutional (CWConv) layer is designed to replace the first convolutional layer of the standard CNN. This enables the first CWConv layer to discover more meaningful filters. Furthermore, only the scale parameter and translation parameter are directly learned from raw data at this CWConv layer. This provides a very effective way to obtain a customized filter bank, specifically tuned for extracting defect-related impact component embedded in the vibration signal. In addition, three experimental verification using data from laboratory environment are carried out to verify effectiveness of the proposed method for mechanical fault diagnosis. The results show the importance of the designed CWConv layer and the output of CWConv layer is interpretable. Besides, it is found that WKN has fewer parameters, higher fault classification accuracy and faster convergence speed than standard CNN.

cs.CV↗