Deploying humanoid robots in real-world environments requires tightly integrating perception, locomotion, and manipulation under partial observations, dynamic changes, and frequent transitions across heterogeneous sub-tasks. Towards this challenge, we propose EgoActing, a new task that grounds high-level instructions into diverse, precise, and spatially aware humanoid actions. We instantiate it with EgoActor, a hierarchical humanoid system centered on a vision-language model that predicts egocentric textual actions from monocular RGB observations. Among the actions, parameterized movement and viewpoint actions are mapped to locomotion and head controllers, while manipulation and human-interaction actions are routed to downstream skills via lightweight parsing. Trained with broad supervision from real-world egocentric demonstrations, spatial reasoning QA, and simulated demonstrations, EgoActor makes robust, context-aware decisions with a sub-second VLM forward pass for both 8B and 4B variants, while the current cloud-deployed robot loop takes 3–8 seconds end-to-end. Extensive simulated and real-world evaluations show that EgoActor bridges abstract task planning and concrete motor execution, generalizing across diverse tasks and unseen environments. Code and models are released at https://baai-agents.github.io/RoboNoid/EgoActor/.

Visual grounding (VG) aims at locating the foreground entities that match the given natural language expression. Previous datasets and methods for classic VG task mainly rely on the prior assumption that the given expression must literally refer to the target object, which greatly impedes the practical deployment of agents in real-world scenarios. Since users usually prefer to provide the intention-based expressions for the desired object instead of covering all the details, it is necessary for the agents to interpret the intention-driven instructions. Thus, in this work, we take a step further to the intention-driven visual-language (V-L) understanding. To promote classic VG towards human intention interpretation, we propose a new intention-driven visual grounding (IVG) task and build a largest-scale IVG dataset named IntentionVG with free-form intention expressions. Considering that practical agents need to move and find specific targets among various scenarios to realize the grounding task, our IVG task and IntentionVG dataset have taken the crucial properties of both multi-scenario perception and egocentric view into consideration. Besides, various types of models are set up as the baselines to realize our IVG task. Extensive experiments on our IntentionVG dataset and baselines demonstrate the necessity and efficacy of our method for the V-L field. To foster future research in this direction, our newly built dataset and baselines will be publicly available at https://github.com/Rubics-Xuan/IVG.

Dense retrieval is widely used for entity linking to retrieve entities from large-scale knowledge bases. Mainstream techniques are based on a dual-encoder framework, which encodes mentions and entities independently and calculates their relevances via rough interaction metrics, resulting in difficulty in explicitly modeling multiple mention-relevant parts within entities to match divergent mentions. Aiming at learning entity representations that can match divergent mentions, this paper proposes a Multi-View Enhanced Distillation (MVD) framework, which can effectively transfer knowledge of multiple fine-grained and mention-relevant parts within entities from cross-encoders to dual-encoders. Each entity is split into multiple views to avoid irrelevant information being over-squashed into the mention-relevant view. We further design cross-alignment and self-alignment mechanisms for this framework to facilitate fine-grained knowledge distillation from the teacher model to the student model. Meanwhile, we reserve a global-view that embeds the entity as a whole to prevent dispersal of uniform information. Experiments show our method achieves state-of-the-art performance on several entity linking benchmarks.