5 papers
Harnessing Domain Specialists in Multimodal Mixture-of-Experts for Efficient Adaptation
Damiano Marsili, Raphi Kang, Aditya Mehta +2
Mixture-of-Experts (MoE) architectures scale model capacity through sparse computation, routing each token through only a small subset of experts. In this work, we explore whether…
No Labels, No Problem: Training Visual Reasoners with Multimodal Verifiers
Damiano Marsili, Georgia Gkioxari
Visual reasoning is challenging, requiring both precise object grounding and understanding complex spatial relationships. Existing methods fall into two camps: language-only chain-…
Same or Not? Enhancing Visual Perception in Vision-Language Models
Damiano Marsili, Aditya Mehta, Ryan Y. Lin +1
Vision-language models (VLMs) excel at broad visual understanding but remain coarse-grained, exhibit visual biases, and miss subtle visual details. Existing training corpora reinfo…
NeurIPS 2025 E2LM Competition : Early Training Evaluation of Language Models
Mouadh Yagoubi, Yasser Dahou, Billel Mokeddem +12
Existing benchmarks have proven effective for assessing the performance of fully trained large language models. However, we find striking differences in the early training stages o…
Visual Agentic AI for Spatial Reasoning with a Dynamic API
Damiano Marsili, Rohun Agrawal, Yisong Yue +1
Visual reasoning -- the ability to interpret the visual world -- is crucial for embodied agents that operate within three-dimensional scenes. Progress in AI has led to vision and l…