Beyond the Current Scene: Event-Referential Grasping with Active View Selection
Summary
The paper introduces BeyondSCe, a zero-shot robotic grasping system that identifies objects referenced by past human-object interactions rather than names or appearance, and actively selects camera viewpoints to reveal occluded targets, achieving 76-77% grasp success rates that significantly outperform baselines.
View Cached Full Text
Cached at: 10/02/26, 04:26 AM
Paper page - Beyond the Current Scene: Event-Referential Grasping with Active View Selection
Source: https://huggingface.co/papers/2609.39375
Abstract
Arobotthatobservespeopleinteractingwithobjectsshouldbeabletocarryoutlaterrequeststhatreferbacktothoseinteractions.Suchrequestsmayspecifyagrasptargetbytheroleitplayedinapasteventratherthanbyitsnameorappearance.Moreover,thetargetmaynolongerbevisiblewhentherobotisaskedtoact.WepresentBeyondSCe,azero-shotroboticgraspingsystemforthisevent-referentialsetting.Giventheeventhistoryandthecurrentscene,thesystemidentifiestherequestedobjectorpartandlocalizesitforgrasping.Ifthetargetisoccluded,itcombinesaneventpriorrecoveredfromthehistorywithcurrentscenegeometrytoselectcameraviewpointslikelytorevealthetarget.Thesystemusespretrainedmodelswithoutadditionaltask-specifictraining.Inreal-robotexperimentswithasinglewrist-mountedRGB-Dcamera,itachievesgraspsuccessratesof76%and77%forinitiallyvisibleandoccludedtargets,respectively,comparedwith40%and55%forthestrongestbaselineineachcondition.Onfouradditionalsceneswithheavyocclusion,itincreasesgraspsuccessratesfrom75%to95%whilereducingthemeannumberofviewsfrom3.35to2.20,comparedwithanactive-perceptionbaselinegiventhetarget’sground-truth3Dboundingbox.
View arXiv pageView PDFProject pageGitHub4Add to collection
Get this paper in your agent:
hf papers read 2609\.39375
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2609.39375 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2609.39375 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2609.39375 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
SeededGrasp: Language-Guided Grasping in Complex Scenes with Multiple Embodiments
SeededGrasp proposes a data-efficient framework that uses a vision-language model to predict a seed point for a lightweight grasp generator, enabling language-guided grasping in complex scenes with multiple robot embodiments. The method outperforms baselines with 72% simulation and 78% real-world success, and includes a new large-scale multi-embodiment grasping dataset.
Active Perception for Embodied Disambiguation
This paper proposes an active-perception framework for embodied target disambiguation, using vision-language models to decide based on accumulated visual evidence and interaction information. Real-robot experiments demonstrate its effectiveness in combining physical observation with user-intent clarification.
Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping
This paper introduces Agentic RAG-VLM, a unified framework that integrates retrieval-augmented generation with vision-language models and self-reflective planning for generalizable robotic grasping in cluttered environments, achieving 78.3% success rate.
Human Universal Grasping
A flow-matching model generates diverse human grasps from RGB-D images, enabling zero-shot robotic grasping with improved performance over existing methods. The model, trained on a large egocentric dataset, significantly outperforms state-of-the-art baselines on a new benchmark.
Visual-Seeker: Towards Visual-Native Multimodal Agentic Search via Active Visual Reasoning
Visual-Seeker proposes a visual-native multimodal deep search agent that actively reasons over fine-grained visual details and synthesizes multimodal evidence, achieving state-of-the-art performance on five challenging multimodal search benchmarks.