← Back to all news

Fine-tuning and adaptation

LoRA, PEFT, instruction tuning and domain adaptation — adapting a pretrained model without paying to train one.

8,205 stories · RSS feed

arXiv AI
Aug 25

From Isolation to Alignment: Unified LoRA for Efficient Multi-Task Learning

The paper introduces Align‑LoRA, a unified LoRA framework for multi‑task learning that replaces complex, isolated adapter designs with a single‑adapter model enhanced by a higher rank and an explicit alignment loss. It demonstrates that a router‑free, multi‑head model with high inter‑head redundancy can outperform more elaborate baselines, and that a unified LoRA can achieve competitive performance while enabling weight merging and zero inference latency. Extensive experiments and theoretical analysis confirm that Align‑LoRA surpasses prevailing approaches, offering a simpler, production‑friendly paradigm for parameter‑efficient fine‑tuning of large language models.

By Jinda Liu, Yi Chang, Yuan Wu
llmsdiffusionfine-tuningefficiencysafety
More like this →
arXiv Computer Vision
Aug 25

SVD-Based Typicality Maps for Out-of-Distribution Detection in Vision Transformers

The paper introduces a technique for examining Vision Transformers by decomposing each affine layer’s weight matrix with Singular Value Decomposition and projecting activations onto the leading right singular vectors, yielding compact, layer‑intrinsic representations. By fitting class‑conditional density models at each layer, the authors generate per‑class typicality scores that are stacked into two‑dimensional typicality maps, summarizing how class‑specific evidence evolves through the network. From these maps, two post‑hoc out‑of‑distribution detection scores are derived: the Prototype Alignment Score (PAS), which measures agreement with class reference prototypes, and the Multi‑Layer Soft Voting (MLSV) score, which captures cross‑layer consensus without stored prototypes, achieving competitive performance on ViT‑B/16 fine‑tuned on CIFAR‑100 without retraining or OOD exposure.

By Aldo Sean Sartor, Leandro de Souza Rosa, Andriy Enttsel, Mauro Mangia, Riccardo Rovatti
llmsfine-tuningsafety
More like this →
arXiv Computer Vision
Aug 25

Controllable blind deblurring with diffusion models

The paper introduces SuperSharpen, a diffusion-based method for blind deblurring in professional photography that can invert unknown isotropic blur without knowing the degradation kernel. It offers explicit control over restoration strength via a blur measure and compares two conditioning strategies: a ControlNet-style adapter on a frozen backbone and full finetuning of the diffusion prior. Experiments on synthetic and real-world blur show that finetuning yields higher fidelity with fewer hallucinated details, improving perceptual quality and controllable restoration strength.

By Imane Si Salah, Emile Cribelier, Thomas Veit, Wolf Hauser, Arthur Leclaire
diffusionfine-tuning
More like this →
arXiv AI
Aug 25

DiaRelay: Relaying Dialogue Context with a Constant-Size Memory for Emotion Recognition in Conversation

DiaRelay introduces a lightweight adapter that lets large language models maintain a constant‑size dialogue‑level memory for emotion recognition in conversation. It builds on LoRA by adding a Selective Relay Memory Transition that aggregates useful historical evidence into a bounded memory, and a Dual‑axis Relay Memory Read that uses this memory to modulate low‑rank feature transformations. Experiments show DiaRelay achieves state‑of‑the‑art weighted F1 and accuracy on MELD with only 7.1 M additional trainable parameters, while also performing competitively on IEMOCAP.

By Zihao Zhou, Bin Yang, Jinghui Qin, Kebing Jin
llmsfine-tuningbenchmarks
More like this →
arXiv Computer Vision
Aug 25

Does a Modern-Handwriting Warm-Up Help Historical Arabic OCR? A Reproducible, Compute-Matched Evaluation on Muharaf and KHATT

arXiv:2608.22316v1 Announce Type: cross Abstract: Whether an intermediate stage of modern Arabic handwriting helps or hurts historical Arabic HTR is usually decided from one implementation and one co...

By Sumaih Almarshad, Maram Alamri, Dona Aloraini, Fares Altuwaim, AlJawharh AlOtaibi, Reem Alyabis, Rayah Aldawsari
computer-visionfine-tuningmultimodalbenchmarkssafety
More like this →