Large language model editing methods frequently suffer from overfitting, wherein factual updates can propagate beyond their intended scope, overemphasizing the edited target even when it's contextually inappropriate.To address this challenge, we introduce REACT (Representation Extraction And Controllable Tuning), a unified two-phase framework designed for precise and controllable knowledge editing.In the initial phase, we utilize tailored stimuli to extract latent factual representations and apply Principal Component Analysis with a simple learnbale linear transformation to compute a directional "belief shift" vector for each instance.In the second phase, we apply controllable perturbations to hidden states using the obtained vector with a magnitude scalar, gated by a pre-trained classifier that permits edits only when contextually necessary.Relevant experiments on EVOKE benchmarks demonstrate that REACT significantly reduces overfitting across nearly all evaluation metrics, and experiments on COUNTERFACT and MQuAKE shows that our method preserves balanced basic editing performance (reliability, locality, and generality) under diverse editing scenarios.

Protein phosphorylation, a key post-translational modification, provides essential insight into protein properties, making its prediction highly significant. Using the emerging capabilities of large language models (LLMs), we apply Low-Rank Adaptation (LoRA) fine-tuning to ESM2, a powerful protein large language model, to efficiently extract features with minimal computational resources, optimizing task-specific text alignment. Additionally, we integrate the conformer architecture with the feature coupling unit to enhance local and global feature exchange, further improving prediction accuracy. Our model achieves state-of-the-art performance, obtaining area under the curve scores of 79.5%, 76.3%, and 71.4% at the S, T, and Y sites of the general data sets. Based on the powerful feature extraction capabilities of LLMs, we conduct a series of analyses on protein representations, including studies on their structure, sequence, and various chemical properties [such as hydrophobicity (GRAVY), surface charge, and isoelectric point]. We propose a test method called linear regression tomography which is a top-down method using representation to explore the model's feature extraction capabilities. Our resources, including data and code, are publicly accessible at https://github.com/SkywalkerLuke/PhosF3C.

Recently, knowledge editing on large language models (LLMs) has received considerable attention. Compared to this, editing Large Vision-Language Models (LVLMs) faces extra challenges from diverse data modalities and complicated model components, and data for LVLMs editing are limited. The existing LVLM editing benchmark, which comprises three metrics (Reliability, Locality, and Generality), falls short in the quality of synthesized evaluation images and cannot assess whether models apply edited knowledge in relevant content. Therefore, we employ more reliable data collection methods to construct a new Large $\textbf{V}$ision-$\textbf{L}$anguage Model $\textbf{K}$nowledge $\textbf{E}$diting $\textbf{B}$enchmark, $\textbf{VLKEB}$, and extend the Portability metric for more comprehensive evaluation. Leveraging a multi-modal knowledge graph, our image data are bound with knowledge entities. This can be further used to extract entity-related knowledge, which constitutes the base of editing data. We conduct experiments of different editing methods on five LVLMs, and thoroughly analyze how do they impact the models. The results reveal strengths and deficiencies of these methods and hopefully provide insights for future research. The codes and dataset are available at: https://github.com/VLKEB/VLKEB.

Phosphorylation is pivotal in numerous fundamental cellular processes and plays a significant role in the onset and progression of various diseases. The accurate identification of these phosphorylation sites is crucial for unraveling the molecular mechanisms within cells and during viral infections, potentially leading to the discovery of novel therapeutic targets. In this study, we develop PTransIPs, a new deep learning framework for the identification of phosphorylation sites. Independent testing results demonstrate that PTransIPs outperforms existing state-of-the-art (SOTA) methods, achieving AUCs of 0.9232 and 0.9660 for the identification of phosphorylated S/T and Y sites, respectively. PTransIPs contributes from three aspects. 1) PTransIPs is the first to apply protein pre-trained language model (PLM) embeddings to this task. It utilizes ProtTrans and EMBER2 to extract sequence and structure embeddings, respectively, as additional inputs into the model, effectively addressing issues of dataset size and overfitting, thus enhancing model performance; 2) PTransIPs is based on Transformer architecture, optimized through the integration of convolutional neural networks and TIM loss function, providing practical insights for model design and training; 3) The encoding of amino acids in PTransIPs enables it to serve as a universal framework for other peptide bioactivity tasks, with its excellent performance shown in extended experiments of this paper.