action alignment 1behavior cloning 1representation anchoring 1robot manipulation 1vision-language models 1
From the 1 of 3 linked papers with an AI index.
3 papers
cs.RO2026
Generalizable VLA Finetuning via Representation Anchoring and Language-Action Alignment
Dwip Dalal, Shivansh Patel, Chahit Jain +7
The paper introduces Anchor-Align, a method that adds representation anchoring and language-action alignment to behavior‑cloning finetuning of vision‑language models for robot mani…
cs.CV2026
Constructive Distortion: Improving MLLMs with Attention-Guided Image Warping
Dwip Dalal, Gautam Vashishtha, Utkarsh Mishra +6
Multimodal large language models (MLLMs) often miss small details and spatial relations in cluttered scenes, leading to errors in fine-grained perceptual grounding. We introduce At…
cs.CV2026
City Navigation in the Wild: Exploring Emergent Navigation from Web-Scale Knowledge in MLLMs
Dwip Dalal, Utkarsh Mishra, Narendra Ahuja +1
Leveraging multimodal large language models (MLLMs) to develop embodied agents offers significant promise for addressing complex real-world tasks. However, current evaluation bench…