English
Related papers

Related papers: Cross-Architecture Model Diffing with Crosscoders:…

200 papers

Model diffing is the study of how fine-tuning changes a model's representations and internal algorithms. Many behaviors of interest are introduced during fine-tuning, and model diffing offers a promising lens to interpret such behaviors.…

Machine Learning · Computer Science 2026-02-23 Julian Minder , Clément Dumas , Caden Juang , Bilal Chugtai , Neel Nanda

Model diffing methods aim to identify how fine-tuning changes a model's internal representations. Crosscoders approach this by learning shared dictionaries of interpretable latent directions between base and fine-tuned models. However,…

Machine Learning · Computer Science 2026-03-06 Aly Kassem , Thomas Jiralerspong , Negar Rostamzadeh , Golnoosh Farnadi

Backdoor attacks on language models pose a significant threat to AI safety, where models behave normally on most inputs but exhibit harmful behavior when triggered by specific patterns. Detecting such backdoors through mechanistic…

Computation and Language · Computer Science 2026-05-11 Sachin Kumar

Multilingual Large Language Models (LLMs) can process many languages, yet how they internally represent this diversity remains unclear. Do they form shared multilingual representations with language-specific decoding, and if so, why does…

Computation and Language · Computer Science 2026-02-10 Abir Harrasse , Florent Draye , Punya Syon Pandey , Zhijing Jin , Bernhard Schölkopf

Institutions with limited data and computing resources often outsource model training to third-party providers in a semi-honest setting, assuming adherence to prescribed training protocols with pre-defined learning paradigm (e.g.,…

Machine Learning · Computer Science 2025-04-02 Xuan Wang , Siyuan Liang , Dongping Liao , Han Fang , Aishan Liu , Xiaochun Cao , Yu-liang Lu , Ee-Chien Chang , Xitong Gao

Large language models (LLMs) learn non-trivial abstractions during pretraining, such as detecting irregular plural noun subjects. However, because traditional evaluation methods (e.g., benchmarking) fail to reveal how models acquire these…

Computation and Language · Computer Science 2026-05-01 Deniz Bayazit , Aaron Mueller , Antoine Bosselut

Being able to identify functions of interest in cross-architecture software is useful whether you are analysing for malware, securing the software supply chain or conducting vulnerability research. Cross-Architecture Binary Code Similarity…

Cryptography and Security · Computer Science 2023-11-30 Josh Collyer , Tim Watson , Iain Phillips

The use of large language models (LLMs) in qualitative analysis offers enhanced efficiency but raises questions about their alignment with the contextual nature of research for design (RfD). This research examines the trustworthiness of…

Human-Computer Interaction · Computer Science 2025-04-24 Joel Oksanen , Andrés Lucero , Perttu Hämäläinen

Diffusion large language models (dLLMs) are compelling alternatives to autoregressive (AR) models because their denoising models operate over the entire sequence. The global planning and iterative refinement features of dLLMs are…

Computation and Language · Computer Science 2025-06-27 Shansan Gong , Ruixiang Zhang , Huangjie Zheng , Jiatao Gu , Navdeep Jaitly , Lingpeng Kong , Yizhe Zhang

Software clones are beneficial to detect security gaps and software maintenance in one programming language or across multiple languages. The existing work on source clone detection performs well but in a single programming language.…

Software Engineering · Computer Science 2022-05-11 Mohammad A. Yahya , Dae-Kyoo Kim

Many features in pretrained Transformers span multiple layers: they emerge through stages of inference, persist in the residual stream, or are built jointly by parallel MLPs. Crosscoders (namely, sparse dictionaries trained jointly across…

Machine Learning · Computer Science 2026-05-12 Andreas D. Demou , Panagiotis Koromilas , James Oldfield , Yannis Panagakis , Mihalis A. Nicolaou

Deepfake technology poses a significant threat to security and social trust. Although existing detection methods have shown high performance in identifying forgeries within datasets that use the same deepfake techniques for both training…

Computer Vision and Pattern Recognition · Computer Science 2024-10-22 Shanmin Yang , Hui Guo , Shu Hu , Bin Zhu , Ying Fu , Siwei Lyu , Xi Wu , Xin Wang

Crack Segmentation in industrial concrete surfaces is a challenging task because cracks usually exhibit intricate morphology with slender appearances. Traditional segmentation methods often struggle to accurately locate such cracks, leading…

Computer Vision and Pattern Recognition · Computer Science 2025-01-23 Xianglong Shi , Yunhan Jiang , Xiaoheng Jiang , Mingling Xu , Yang Liu

Deep learning-based recognition systems are deployed at scale for several real-world applications that inevitably involve our social life. Although being of great support when making complex decisions, they might capture spurious data…

Computer Vision and Pattern Recognition · Computer Science 2023-03-28 Leonardo Iurada , Silvia Bucci , Timothy M. Hospedales , Tatiana Tommasi

Large language models (LLMs) show strong potential for neural architecture generation, yet existing approaches produce complete model implementations from scratch -- computationally expensive and yielding verbose code. We propose Delta-Code…

Machine Learning · Computer Science 2026-05-07 Santosh Premi Adhikari , Radu Timofte , Dmitry Ignatov

Large language models (LLMs) are increasingly used to assign document relevance labels in information retrieval pipelines, especially in domains lacking human-labeled data. However, different models often disagree on borderline cases,…

Information Retrieval · Computer Science 2025-07-04 William A. Ingram , Bipasha Banerjee , Edward A. Fox

Production language-model systems answer a request by partitioning it across an invisible orchestration of worker agents that recompose one integrated report. We ask what this does to a class of defect no single worker can see: a…

Software Engineering · Computer Science 2026-05-27 Hiroki Fukui

The cross-depiction problem is that of recognising visual objects regardless of whether they are photographed, painted, drawn, etc. It is a potentially significant yet under-researched problem. Emulating the remarkable human ability to…

Computer Vision and Pattern Recognition · Computer Science 2015-05-04 Hongping Cai , Qi Wu , Tadeo Corradi , Peter Hall

Recent advances in large-scale code generation models have led to remarkable progress in producing high-quality code. These models are trained in a self-supervised manner on extensive unlabeled code corpora using a decoder-only…

Software Engineering · Computer Science 2026-02-12 Jiayi Lin , Yanlin Wang , Yibiao Yang , Lei Zhang , Yutao Xie

With the rapid adoption of large language models (LLMs) in automated code refactoring, assessing and ensuring functional equivalence between LLM-generated refactoring and the original implementation becomes critical. While prior work…

Software Engineering · Computer Science 2026-02-18 Simantika Bhattacharjee Dristi , Matthew B. Dwyer
‹ Prev 1 2 3 10 Next ›