The paper investigates whether using fine-grained, tensorized adapters (CP components) instead of standard LoRA ranks improves the accuracy-budget trade-off in PEFT, finding that while they fill budge…

The paper introduces NaRA, a noise-aware LoRA technique that dynamically adapts fine-tuning parameters based on the noise level during diffusion, significantly improving the performance of Diffusion L…

The paper reframes Parameter-Efficient Fine-Tuning (PEFT) from a mere cost-saving alternative to a robust architecture for creating persistent, personalized models that layer specific behaviors onto l…

The paper proposes a unified framework to systematically redefine instance matching for Panoptic Quality evaluation, moving beyond the standard One-to-One matching to accommodate complex scenarios lik…

The paper demonstrates that Low-Rank Adaptation (LoRA) is an effective and superior method for adapting large, pretrained Transformer surrogates for automotive aerodynamics to new vehicle families usi…

KLAS introduces a novel framework that uses KL divergence to automatically select optimal pairs of pretrained models for stitching, significantly improving the accuracy-efficiency tradeoff of resultin…

The paper demonstrates that supervised fine-tuning significantly outperforms frontier zero-shot large language models for screen-conditioned action prediction on the PiSAR benchmark, highlighting the…

TRACE proposes a novel method to mitigate catastrophic forgetting in continual LLM fine-tuning by identifying and isolating a small, task-specific subset of essential parameters for each task.

The paper introduces a novel, transferable learned attack (LT-MIA) that detects a universal 'signature of memorization' in language models, achieving high accuracy across diverse model architectures (…

The paper quantifies the exact parametric memory capacity of LLMs using LoRA and proposes a new optimization strategy, MemFT, to enhance memory fidelity.

FLORO is a multimodal geospatial foundation model that learns transferable remote sensing representations from a small, diverse corpus, achieving strong performance across various sensor types and res…

LayerRoute introduces a lightweight, input-conditioned adapter that selectively skips transformer blocks in agentic language models, achieving significant FLOPs reduction while improving performance.

The paper introduces Parallax, a scalable and numerically stable parameterized Local Linear Attention mechanism that significantly improves LLM performance and efficiency compared to existing methods…

This paper demonstrates that adversarial perturbations possess a low-rank structure, and proposes a two-step method to leverage this property to significantly improve the efficiency and effectiveness…

ProtoAda introduces a prototype-guided, format-aware adaptive tuning framework to improve multimodal continual instruction tuning by ensuring task assignment and parameter updates respect heterogeneou…