Xiangtong Yao

Xiangtong Yao, Hongkuan Zhou, Oier Mees et al. · cmu

Language-conditioned robot manipulation is an emerging field aimed at enabling seamless communication and cooperation between humans and robotic agents by teaching robots to comprehend and execute instructions conveyed in natural language. This interdisciplinary area integrates scene understanding, language processing, and policy learning to bridge the gap between human instructions and robot actions. In this comprehensive survey, we systematically explore recent advancements in language-conditioned robot manipulation. We categorize existing methods based on the primary ways language is integrated into the robot system, namely language for state evaluation, language as a policy condition, language for cognitive planning and reasoning, and language in unified vision-language-action models. Specifically, we further analyze state-of-the-art techniques from five axes of action granularity, data and supervision regimes, system cost and latency, environments and evaluations, and cross-modal task specification. Additionally, we highlight the key debates in the field. Finally, we discuss open challenges and future research directions, focusing on potentially enhancing generalization capabilities and addressing safety issues in language-conditioned robot manipulators.

Xiangtong Yao, Zhenshan Bing, Genghang Zhuang et al.

Meta-reinforcement learning (meta-RL) is a promising approach that enables the agent to learn new tasks quickly. However, most meta-RL algorithms show poor generalization in multi-task scenarios due to the insufficient task information provided only by rewards. Language-conditioned meta-RL improves the generalization capability by matching language instructions with the agent's behaviors. While both behaviors and language instructions have symmetry, which can speed up human learning of new knowledge. Thus, combining symmetry and language instructions into meta-RL can help improve the algorithm's generalization and learning efficiency. We propose a dual-MDP meta-reinforcement learning method that enables learning new tasks efficiently with symmetrical behaviors and language instructions. We evaluate our method in multiple challenging manipulation tasks, and experimental results show that our method can greatly improve the generalization and learning efficiency of meta-reinforcement learning. Videos are available at https://tumi6robot.wixsite.com/symmetry/.

Mingyang Wang, Zhenshan Bing, Xiangtong Yao et al.

Meta-reinforcement learning enables artificial agents to learn from related training tasks and adapt to new tasks efficiently with minimal interaction data. However, most existing research is still limited to narrow task distributions that are parametric and stationary, and does not consider out-of-distribution tasks during the evaluation, thus, restricting its application. In this paper, we propose MoSS, a context-based Meta-reinforcement learning algorithm based on Self-Supervised task representation learning to address this challenge. We extend meta-RL to broad non-parametric task distributions which have never been explored before, and also achieve state-of-the-art results in non-stationary and out-of-distribution tasks. Specifically, MoSS consists of a task inference module and a policy module. We utilize the Gaussian mixture model for task representation to imitate the parametric and non-parametric task variations. Additionally, our online adaptation strategy enables the agent to react at the first sight of a task change, thus being applicable in non-stationary tasks. MoSS also exhibits strong generalization robustness in out-of-distributions tasks which benefits from the reliable and robust task representation. The policy is built on top of an off-policy RL algorithm and the entire network is trained completely off-policy to ensure high sample efficiency. On MuJoCo and Meta-World benchmarks, MoSS outperforms prior works in terms of asymptotic performance, sample efficiency (3-50x faster), adaptation efficiency, and generalization robustness on broad and diverse task distributions.

Ziou Hu, Xiangtong Yao, Yuan Meng et al.

Diffusion policies excel at visuomotor control but often fail catastrophically under severe out-of-distribution (OOD) disturbances, such as unexpected object displacements or visual corruptions. To address this vulnerability, we introduce the Dream Diffusion Policy (DDP), a framework that deeply integrates a diffusion world model into the policy's training objective via a shared 3D visual encoder. This co-optimization endows the policy with robust state-prediction capabilities. When encountering sudden OOD anomalies during inference, DDP detects the real-imagination discrepancy and actively abandons the corrupted visual stream. Instead, it relies on its internal "imagination" (autoregressively forecasted latent dynamics) to safely bypass the disruption, generating imagined trajectories before smoothly realigning with physical reality. Extensive evaluations demonstrate DDP's exceptional resilience. Notably, DDP achieves a 73.8% OOD success rate on MetaWorld (vs. 23.9% without predictive imagination) and an 83.3% success rate under severe real-world spatial shifts (vs. 3.3% without predictive imagination). Furthermore, as a stress test, DDP maintains a 76.7% real-world success rate even when relying entirely on open-loop imagination post-initialization.

Dingyi Zhou, Mu He, Zhuowei Fang et al.

We introduce AffordanceGrasp-R1, a reasoning-driven affordance segmentation framework for robotic grasping that combines a chain-of-thought (CoT) cold-start strategy with reinforcement learning to enhance deduction and spatial grounding. In addition, we redesign the grasping pipeline to be more context-aware by generating grasp candidates from the global scene point cloud and subsequently filtering them using instruction-conditioned affordance masks. Extensive experiments demonstrate that AffordanceGrasp-R1 consistently outperforms state-of-the-art (SOTA) methods on benchmark datasets, and real-world robotic grasping evaluations further validate its robustness and generalization under complex language-conditioned manipulation scenarios.

Zebin Jiang, Tianle Jin, Xiangtong Yao et al.

Grasping is one of the most fundamental challenging capabilities in robotic manipulation, especially in unstructured, cluttered, and semantically diverse environments. Recent researches have increasingly explored language-guided manipulation, where robots not only perceive the scene but also interpret task-relevant natural language instructions. However, existing language-conditioned grasping methods typically rely on shallow fusion strategies, leading to limited semantic grounding and weak alignment between linguistic intent and visual grasp reasoning.In this work, we propose Language-Guided Grasp Detection (LGGD) with a coarse-to-fine learning paradigm for robotic manipulation. LGGD leverages CLIP-based visual and textual embeddings within a hierarchical cross-modal fusion pipeline, progressively injecting linguistic cues into the visual feature reconstruction process. This design enables fine-grained visual-semantic alignment and improves the feasibility of the predicted grasps with respect to task instructions. In addition, we introduce a language-conditioned dynamic convolution head (LDCH) that mixes multiple convolution experts based on sentence-level features, enabling instruction-adaptive coarse mask and grasp predictions. A final refinement module further enhances grasp consistency and robustness in complex scenes.Experiments on the OCID-VLG and Grasp-Anything++ datasets show that LGGD surpasses existing language-guided grasping methods, exhibiting strong generalization to unseen objects and diverse language queries. Moreover, deployment on a real robotic platform demonstrates the practical effectiveness of our approach in executing accurate, instruction-conditioned grasp actions. The code will be released publicly upon acceptance.

Yu Zhang, Guangyao Tian, Long Wen et al.

This paper proposes a LiDAR-based goal-seeking and exploration framework, addressing the efficiency of online obstacle avoidance in unstructured environments populated with static and moving obstacles. This framework addresses two significant challenges associated with traditional dynamic control barrier functions (D-CBFs): their online construction and the diminished real-time performance caused by utilizing multiple D-CBFs. To tackle the first challenge, the framework's perception component begins with clustering point clouds via the DBSCAN algorithm, followed by encapsulating these clusters with the minimum bounding ellipses (MBEs) algorithm to create elliptical representations. By comparing the current state of MBEs with those stored from previous moments, the differentiation between static and dynamic obstacles is realized, and the Kalman filter is utilized to predict the movements of the latter. Such analysis facilitates the D-CBF's online construction for each MBE. To tackle the second challenge, we introduce buffer zones, generating Type-II D-CBFs online for each identified obstacle. Utilizing these buffer zones as activation areas substantially reduces the number of D-CBFs that need to be activated. Upon entering these buffer zones, the system prioritizes safety, autonomously navigating safe paths, and hence referred to as the exploration mode. Exiting these buffer zones triggers the system's transition to goal-seeking mode. We demonstrate that the system's states under this framework achieve safety and asymptotic stabilization. Experimental results in simulated and real-world environments have validated our framework's capability, allowing a LiDAR-equipped mobile robot to efficiently and safely reach the desired location within dynamic environments containing multiple obstacles.

Lin Hong, Xiangtong Yao, Mürüvvet Bozkurt et al.

Underwater salient instance segmentation (USIS) is crucial for marine robotic systems, as it enables both underwater salient object detection and instance-level mask prediction for visual scene understanding. Compared with its terrestrial counterpart, USIS is more challenging due to the underwater image degradation. To address this issue, this paper proposes USIS-PGM, a single-stage framework for USIS. Specifically, the encoder enhances boundary cues through a frequency-aware module and performs content-adaptive feature reweighting via a dynamic weighting module. The decoder incorporates a Transformer-based instance activation module to better distinguish salient instances. In addition, USIS-PGM employs multi-scale Gaussian heatmaps generated from ground-truth masks through Photometric Gaussian Mixture (PGM) to supervise intermediate decoder features, thereby improving salient instance localization and producing more structurally coherent mask predictions. Experimental results demonstrate the superiority and practical applicability of the proposed USIS-PGM model.

Yu Zhang, Long Wen, Xiangtong Yao et al.

This paper presents an adaptive online learning framework for systems with uncertain parameters to ensure safety-critical control in non-stationary environments. Our approach consists of two phases. The initial phase is centered on a novel sparse Gaussian process (GP) framework. We first integrate a forgetting factor to refine a variational sparse GP algorithm, thus enhancing its adaptability. Subsequently, the hyperparameters of the Gaussian model are trained with a specially compound kernel, and the Gaussian model's online inferential capability and computational efficiency are strengthened by updating a solitary inducing point derived from new samples, in conjunction with the learned hyperparameters. In the second phase, we propose a safety filter based on high-order control barrier functions (HOCBFs), synergized with the previously trained learning model. By leveraging the compound kernel from the first phase, we effectively address the inherent limitations of GPs in handling high-dimensional problems for real-time applications. The derived controller ensures a rigorous lower bound on the probability of satisfying the safety specification. Finally, the efficacy of our proposed algorithm is demonstrated through real-time obstacle avoidance experiments executed using both a simulation platform and a real-world 7-DOF robot.

Yuan Meng, Xiangtong Yao, Haihui Ye et al.

Embodied long-horizon manipulation requires robotic systems to process multimodal inputs-such as vision and natural language-and translate them into executable actions. However, existing learning-based approaches often depend on large, task-specific datasets and struggle to generalize to unseen scenarios. Recent methods have explored using large language models (LLMs) as high-level planners that decompose tasks into subtasks using natural language and guide pretrained low-level controllers. Yet, these approaches assume perfect execution from low-level policies, which is unrealistic in real-world environments with noise or suboptimal behaviors. To overcome this, we fully discard the pretrained low-level policy and instead use the LLM to directly generate executable code plans within a closed-loop framework. Our planner employs chain-of-thought (CoT)-guided few-shot learning with incrementally structured examples to produce robust and generalizable task plans. Complementing this, a reporter evaluates outcomes using RGB-D and delivers structured feedback, enabling recovery from misalignment and replanning under partial observability. This design eliminates per-step inference, reduces computational overhead, and limits error accumulation that was observed in previous methods. Our framework achieves state-of-the-art performance on 30+ diverse seen and unseen long-horizon tasks across LoHoRavens, CALVIN, Franka Kitchen, and cluttered real-world settings.

Hongkuan Zhou, Stefan Schmid, Yicong Li et al.

The autonomous driving field has seen remarkable advancements in various topics, such as object recognition, trajectory prediction, and motion planning. However, current approaches face limitations in effectively comprehending the complex evolutions of driving scenes over time. This paper proposes FM4SU, a novel methodology for training a symbolic foundation model (FM) for scene understanding in autonomous driving. It leverages knowledge graphs (KGs) to capture sensory observation along with domain knowledge such as road topology, traffic rules, or complex interactions between traffic participants. A bird's eye view (BEV) symbolic representation is extracted from the KG for each driving scene, including the spatio-temporal information among the objects across the scenes. The BEV representation is serialized into a sequence of tokens and given to pre-trained language models (PLMs) for learning an inherent understanding of the co-occurrence among driving scene elements and generating predictions on the next scenes. We conducted a number of experiments using the nuScenes dataset and KG in various scenarios. The results demonstrate that fine-tuned models achieve significantly higher accuracy in all tasks. The fine-tuned T5 model achieved a next scene prediction accuracy of 86.7%. This paper concludes that FM4SU offers a promising foundation for developing more comprehensive models for scene understanding in autonomous driving.

Yuan Meng, Xiangtong Yao, Kejia Chen et al.

Reinforcement learning (RL) methods typically learn new tasks from scratch, often disregarding prior knowledge that could accelerate the learning process. While some methods incorporate previously learned skills, they usually rely on a fixed structure, such as a single Gaussian distribution, to define skill priors. This rigid assumption can restrict the diversity and flexibility of skills, particularly in complex, long-horizon tasks. In this work, we introduce a method that models potential primitive skill motions as having non-parametric properties with an unknown number of underlying features. We utilize a Bayesian non-parametric model, specifically Dirichlet Process Mixtures, enhanced with birth and merge heuristics, to pre-train a skill prior that effectively captures the diverse nature of skills. Additionally, the learned skills are explicitly trackable within the prior space, enhancing interpretability and control. By integrating this flexible skill prior into an RL framework, our approach surpasses existing methods in long-horizon manipulation tasks, enabling more efficient skill transfer and task success in complex environments. Our findings show that a richer, non-parametric representation of skill priors significantly improves both the learning and execution of challenging robotic tasks. All data, code, and videos are available at https://ghiara.github.io/HELIOS/.

Zhihao Zhao, Yinzheng Zhao, Junjie Yang et al.

Significant advancements in AI-driven multimodal medical image diagnosis have led to substantial improvements in ophthalmic disease identification in recent years. However, acquiring paired multimodal ophthalmic images remains prohibitively expensive. While fundus photography is simple and cost-effective, the limited availability of OCT data and inherent modality imbalance hinder further progress. Conventional approaches that rely solely on fundus or textual features often fail to capture fine-grained spatial information, as each imaging modality provides distinct cues about lesion predilection sites. In this study, we propose a novel unpaired multimodal framework \UOPSL that utilizes extensive OCT-derived spatial priors to dynamically identify predilection sites, enhancing fundus image-based disease recognition. Our approach bridges unpaired fundus and OCTs via extended disease text descriptions. Initially, we employ contrastive learning on a large corpus of unpaired OCT and fundus images while simultaneously learning the predilection sites matrix in the OCT latent space. Through extensive optimization, this matrix captures lesion localization patterns within the OCT feature space. During the fine-tuning or inference phase of the downstream classification task based solely on fundus images, where paired OCT data is unavailable, we eliminate OCT input and utilize the predilection sites matrix to assist in fundus image classification learning. Extensive experiments conducted on 9 diverse datasets across 28 critical categories demonstrate that our framework outperforms existing benchmarks.

Zhihao Zhao, Yinzheng Zhao, Junjie Yang et al.

Recent advancements in foundation models, such as the Segment Anything Model (SAM), have significantly impacted medical image segmentation, especially in retinal imaging, where precise segmentation is vital for diagnosis. Despite this progress, current methods face critical challenges: 1) modality ambiguity in textual disease descriptions, 2) a continued reliance on manual prompting for SAM-based workflows, and 3) a lack of a unified framework, with most methods being modality- and task-specific. To overcome these hurdles, we propose CLIP-unified Auto-Prompt Segmentation (\CLAPS), a novel method for unified segmentation across diverse tasks and modalities in retinal imaging. Our approach begins by pre-training a CLIP-based image encoder on a large, multi-modal retinal dataset to handle data scarcity and distribution imbalance. We then leverage GroundingDINO to automatically generate spatial bounding box prompts by detecting local lesions. To unify tasks and resolve ambiguity, we use text prompts enhanced with a unique "modality signature" for each imaging modality. Ultimately, these automated textual and spatial prompts guide SAM to execute precise segmentation, creating a fully automated and unified pipeline. Extensive experiments on 12 diverse datasets across 11 critical segmentation categories show that CLAPS achieves performance on par with specialized expert models while surpassing existing benchmarks across most metrics, demonstrating its broad generalizability as a foundation model.

Hongkuan Zhou, Zhenshan Bing, Xiangtong Yao et al.

The growing interest in language-conditioned robot manipulation aims to develop robots capable of understanding and executing complex tasks, with the objective of enabling robots to interpret language commands and manipulate objects accordingly. While language-conditioned approaches demonstrate impressive capabilities for addressing tasks in familiar environments, they encounter limitations in adapting to unfamiliar environment settings. In this study, we propose a general-purpose, language-conditioned approach that combines base skill priors and imitation learning under unstructured data to enhance the algorithm's generalization in adapting to unfamiliar environments. We assess our model's performance in both simulated and real-world environments using a zero-shot setting. In the simulated environment, the proposed approach surpasses previously reported scores for CALVIN benchmark, especially in the challenging Zero-Shot Multi-Environment setting. The average completed task length, indicating the average number of tasks the agent can continuously complete, improves more than 2.5 times compared to the state-of-the-art method HULC. In addition, we conduct a zero-shot evaluation of our policy in a real-world setting, following training exclusively in simulated environments without additional specific adaptations. In this evaluation, we set up ten tasks and achieved an average 30% improvement in our approach compared to the current state-of-the-art approach, demonstrating a high generalization capability in both simulated environments and the real world. For further details, including access to our code and videos, please refer to https://hk-zh.github.io/spil/