5 papers
Mixture of Layers: Dynamic Layer Routing for Visual Reasoning
Jeonghwan Kim, Sofia Stoica, Jiwan Chung +5
Pre-trained vision encoders contain layer-wise visual representations that differ in spatial granularity, semantic abstraction, and sensitivity to local details. However, most Mult…
Feature Recovery for Object Understanding After Irreversible Fire Damage
Aditi Tiwari, Sofia Stoica, Savya Khosla +2
Objects in post-fire environments often undergo irreversible physical transformations that change their geometry, material state, and visual appearance. Detecting and identifying t…
AcquisitionSynthesis: Targeted Data Generation using Acquisition Functions
Ishika Agarwal, Sofia Stoica, Emre Can Acikgoz +4
Data quality remains a critical bottleneck in developing capable, competitive models. Researchers have explored many ways to generate top quality samples. Some works rely on reject…
SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
Ayush Barik, Sofia Stoica, Nikhil Sarda +4
Text-to-audio (T2A) diffusion models generate high-quality audio but require tens of neural function evaluations (NFEs), resulting in substantial inference latency. Existing accele…
Perception-Aware Policy Optimization for Multimodal Reasoning
Zhenhailong Wang, Xuehang Guo, Sofia Stoica +8
Reinforcement Learning with Verifiable Rewards (RLVR) has proven to be a highly effective strategy for endowing Large Language Models (LLMs) with robust multi-step reasoning abilit…