analiza aiqbench
This arXiv preprint proposes a training and attention method for language-guided 3D affordance grounding, aimed at improving performance on objects not seen during training. Its central claim is that end-to-end training overfits query representations to seen objects, and that alignment and 2D visual guidance can narrow that gap. Readers tracking generalization in vision-language models should note that the reported gains rest on the LASO and PIAD benchmarks, so robustness beyond them remains untested.
BridgeAfford: Bridging Query-Target Gap for Generalizable 3D Affordance Grounding
arXiv:2610.21918v1 Announce Type: cross Abstract: Language-guided 3D affordance grounding aims to localize functional regions on 3D objects according to interaction intents. However, existing methods exhibit poor generalization to unseen objects. One challenge is the query-target gap: a semantic discrepancy between query features encoding interaction intents and target-region features, leading to inaccurate grounding. End-to-end training favors query representations tailored to seen objects, limiting their compatibility with shifted target-region features on unseen objects. To address this issue, we propose BridgeAfford, a framework that reduces the query-target semantic gap from two complementary perspectives. First, Interaction Region Alignment (IRA) promotes cross-object alignment among geometrically similar interaction regions during training. This enables structures on unseen objects that resemble previously observed ones to produce representations similar to those learned during training, facilitating target-region recognition. Second, Visual Evidence Aggregation (VEA) leverages a 2D affordance model to refine the attention distribution and guide the query toward the correct region, yielding more target-aware queries. Extensive experiments on the LASO and PIAD datasets demonstrate that BridgeAfford outperforms existing methods, particularly on unseen objects, highlighting its robust generalization ability.
Abstract:Language-guided 3D affordance grounding aims to localize functional regions on 3D objects according to interaction intents. However, existing methods exhibit poor generalization to unseen objects. One challenge is the query-target gap: a semantic discrepancy between query features encoding interaction intents and target-region features, leading to inaccurate grounding. End-to-end training favors query representations tailored to seen objects, limiting their compatibility with shifted target-region features on unseen objects. To address this issue, we propose BridgeAfford, a framework that reduces the query-target semantic gap from two complementary perspectives. First, Interaction Region Alignment (IRA) promotes cross-object alignment among geometrically similar interaction regions during training. This enables structures on unseen objects that resemble previously observed ones to produce representations similar to those learned during training, facilitating target-region recognition. Second, Visual Evidence Aggregation (VEA) leverages a 2D affordance model to refine the attention distribution and guide the query toward the correct region, yielding more target-aware queries. Extensive experiments on the LASO and PIAD datasets demonstrate that BridgeAfford outperforms existing methods, particularly on unseen objects, highlighting its robust generalization ability.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2610.21918 [cs.CV]
(or arXiv:2610.21918v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2610.21918 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 273 cuvinte
Subiecte conexe
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · 4 sept. 2026
UniAfford: Token-Routed Multitask Learning for Generalizable 2D-3D Affordance Perception
arXiv · 30 sept. 2026
Academia is for Ambition — Alex Zhang, MIT
Latent Space · 2 oct. 2026
FuncBridge: Towards Functional Tool-Use Generalization via Keypoint Trajectory Reasoning
arXiv · 2 oct. 2026