2 papers
cs.LG2026
GRIT -- Geometry-Aware PEFT with K-FACPreconditioning, Fisher-Guided Reprojection, andDynamic Rank Adaptation
Pritish Saha, Chandrav Rajbangshi, Rudra Goyal +6
Parameter-efficient fine-tuning (PEFT) is the default way to adapt LLMs, but widely used LoRA and QLoRA are largely geometry-agnostic: they optimize in fixed, randomly oriented low…
cs.CV2025
MapTrace: Scalable Data Generation for Route Tracing on Maps
Artemis Panagopoulou, Aveek Purohit, Achin Kulshrestha +2
While Multimodal Large Language Models have achieved human-like performance on many visual and textual reasoning tasks, their proficiency in fine-grained spatial understanding, suc…