Fine-tuning and adaptation

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

8,201 stories · RSS feed

arXiv AI

Oct 27

Multi-Fidelity Training Reshapes Uncertainty Decomposition in Atomistic Models

The study examines how multi‑fidelity training—using abundant low‑fidelity data and scarce high‑fidelity labels—affects uncertainty decomposition in atomistic models. By training deep ensembles of mean‑variance MACE models on ANI‑1x, the authors compare high‑fidelity training from scratch with low‑fidelity pretraining followed by fine‑tuning, evaluating aleatoric, epistemic, and total uncertainty under unseen molecular systems and higher‑energy configurations. Results show that pretraining lowers prediction error and improves uncertainty‑based ranking, but does not produce well‑calibrated predictive variances, especially under energy extrapolation, and that fine‑tuning yields a more effective uncertainty‑based acquisition function.

By Matteo Tolloso, Davide Bacciu, Luigi Bonati

fine-tuning

arXiv AI

Oct 27

Evaluating Harmful Memorization in LLM Code Generation

The paper introduces the Harmful Memorization Index (HMI) to measure when large language models (LLMs) fail to adapt to small semantic changes in coding tasks, producing code that is superficially similar to the original solution. It proposes an automated code‑rewriting pipeline that generates semantically different variants of tasks by altering core logic elements, and uses HMI to combine similarity metrics with performance degradation. Experiments on MBPP+ and BigCodeBench show that HMI does not simply grow with model size, is higher for certain logic‑edit types, and that supervised fine‑tuning can increase HMI while PPO offers a better accuracy–HMI trade‑off.

By Lizhe Zhang, Wentao Chen, Li Zhong, Letian Peng, Zilong Wang, Jingbo Shang

llms reinforcement-learning fine-tuning

arXiv AI

Oct 27

A Causal Perspective on the Role of Training Dynamics for Interpreting Privacy Risks in Code Models

The paper investigates how different types of personally identifiable information (PII) are learned and leaked by large language models for code (LLM4Code). By creating a diverse PII dataset, fine‑tuning models of various scales, and analyzing training dynamics through a structural causal model, the authors find that leakage risk varies by PII type—easy‑to‑learn data such as IP addresses leak more often, while harder types like keys and passwords leak less. The study provides causal evidence that leakage risks are type‑dependent and suggests that defenses should be tailored to specific PII categories and their learnability.

By Hua Yang, Alejandro Velasco, Sen Fang, Bowen Xu, Denys Poshyvanyk

llms fine-tuning safety

arXiv AI

Oct 27

How well are open sourced AI-generated image detection models out-of-the-box: A comprehensive benchmark study

The study evaluates 16 AI‑generated image detection methods in a zero‑shot setting, testing 23 pretrained variants on 12 datasets with 2.6 million images from 291 generators. Results show no single detector consistently outperforms others, with accuracy ranging from 37.5 % to 75.0 %. The study highlights that training data alignment and generator type significantly affect performance, especially against modern commercial models, and identifies systematic failure patterns across datasets.

By Simiao Ren (Neo), Yuchen Zhou (Neo), Xingyu Shen (Neo), Kidus Zewde (Neo), Tommy Duong (Neo), George Huang (Neo), Hatsanai (Neo), Tiangratanakul (Dennis), Tsang (Dennis), Ng (Dennis), En Wei, Jiayu Xue

diffusion fine-tuning benchmarks safety

arXiv AI

Oct 27

Multi-Agentic AI for Fairness-Aware and Accelerated Multi-modal Large Model Inference in Real-world Mobile Edge Networks

The paper proposes a Multi-Agentic AI framework to enable fairness‑aware and accelerated inference of large, heterogeneous multi‑modal language models on mobile edge networks. It introduces a Global Planning Agent, a Prompt Scheduling Agent, and multiple on‑node LM Deployment Agents that jointly optimize prompt routing and model deployment using natural language reasoning over telemetry and historical data. Experiments on a city‑wide testbed show a 42.9% reduction in end‑to‑end latency versus a state‑of‑the‑art baseline and an 80.6% reduction versus a full‑activation baseline, while preserving high fairness (Jain's index 0.925).

By Haiyuan Li, Hari Madhukumar, Shuangyi Yan, Yulei Wu, Dimitra Simeonidou

llms agents nlp fine-tuning multimodal benchmarks safety

arXiv Machine Learning

Oct 27

Beyond Local Accuracy: Benchmarking the Rollout Reliability of PDE Foundation Models

arXiv:2610.31935v1 Announce Type: new Abstract: PDE foundation models enable data-efficient adaptation but are deployed autoregressively, so short-horizon accuracy need not imply reliable simulation....

By Jaewan Park, Seid Koric, Giri Krishnan, Sudhakar Pamidighantam, Surresh Marru

fine-tuning benchmarks

arXiv Machine Learning

Oct 27

DRIVE: Fine-tuning via Data Contribution- and Diversity-aware Weighting with Prior Regularization

arXiv:2610.32501v1 Announce Type: new Abstract: Single-stage fine-tuning is pervasive, yet catastrophic forgetting persists when the source pre-training data is inaccessible. Data weighting affords a...

By Qing Liu, Xinrui Chen, Yi Du, Ou Wu, Weiyao Zhu

fine-tuning