English
Related papers

Related papers: Ethicist: Targeted Training Data Extraction Throug…

200 papers

Language models (LMs) are often used as zero-shot or few-shot classifiers by scoring label words, but they remain fragile to adversarial prompts. Prior work typically optimizes task- or model-specific triggers, making results difficult to…

Computation and Language · Computer Science 2025-12-10 Sampriti Soor , Suklav Ghosh , Arijit Sur

Recent advances on prompt-tuning cast few-shot classification tasks as a masked language modeling problem. By wrapping input into a template and using a verbalizer which constructs a mapping between label space and label word space,…

Computation and Language · Computer Science 2022-01-17 Yinyi Wei , Tong Mo , Yongtao Jiang , Weiping Li , Wen Zhao

For well over a quarter century, detection systems have been driven by models learned from input features collected from real or simulated environments. An artifact (e.g., network event, potential malware sample, suspicious email) is deemed…

Cryptography and Security · Computer Science 2018-04-03 Z. Berkay Celik , Patrick McDaniel , Rauf Izmailov , Nicolas Papernot , Ryan Sheatsley , Raquel Alvarez , Ananthram Swami

Target speaker extraction (TSE) aims to recover a target speaker's speech from a mixture using a reference utterance as a cue. Most TSE systems adopt conditional auto-encoder architectures with one-step inference. Inspired by test-time…

Sound · Computer Science 2026-03-12 Zhenghai You , Ying Shi , Lantian Li , Dong Wang

It is important to collect credible training samples $(x,y)$ for building data-intensive learning systems (e.g., a deep learning system). Asking people to report complex distribution $p(x)$, though theoretically viable, is challenging in…

Machine Learning · Computer Science 2021-02-26 Jiaheng Wei , Zuyue Fu , Yang Liu , Xingyu Li , Zhuoran Yang , Zhaoran Wang

Memorization in large-scale text-to-image diffusion models poses significant security and intellectual property risks, enabling adversarial attribute extraction and the unauthorized reproduction of sensitive or proprietary features. While…

Machine Learning · Computer Science 2026-01-28 Divya Kothandaraman , Jaclyn Pytlarz

Dataset distillation aims to compress training data into fewer examples via a teacher, from which a student can learn effectively. While its success is often attributed to structure in the data, modern neural networks also memorize specific…

Machine Learning · Computer Science 2026-02-23 Freya Behrens , Lenka Zdeborová

Recent efforts leverage knowledge distillation techniques to develop lightweight and practical sentiment analysis models. These methods are grounded in human-written instructions and large-scale user texts. Despite the promising results,…

Computation and Language · Computer Science 2025-11-04 Guangyu Xie , Yice Zhang , Jianzhu Bao , Qianlong Wang , Yang Sun , Bingbing Wang , Ruifeng Xu

Scheduled sampling is an effective method to alleviate the exposure bias problem of neural machine translation. It simulates the inference scene by randomly replacing ground-truth target input tokens with predicted ones during training.…

Computation and Language · Computer Science 2021-07-23 Yijin Liu , Fandong Meng , Yufeng Chen , Jinan Xu , Jie Zhou

Meta-learning can extract an inductive bias from previous learning experience and assist the training of new tasks. It is often realized through optimizing a meta-model with the evaluation loss of task-specific solvers. Most existing…

Machine Learning · Computer Science 2021-12-20 Su Lu , Han-Jia Ye , Le Gan , De-Chuan Zhan

This paper focuses on task-agnostic prompt compression for better generalizability and efficiency. Considering the redundancy in natural language, existing approaches compress prompts by removing tokens or lexical units according to their…

As the scale of training corpora for large language models (LLMs) grows, model developers become increasingly reluctant to disclose details on their data. This lack of transparency poses challenges to scientific evaluation and ethical…

Computation and Language · Computer Science 2025-05-22 Weichao Zhang , Ruqing Zhang , Jiafeng Guo , Maarten de Rijke , Yixing Fan , Xueqi Cheng

Alignment of large language models (LLMs) with human values and preferences, often achieved through fine-tuning based on human feedback, is essential for ensuring safe and responsible AI behaviors. However, the process typically requires…

Computation and Language · Computer Science 2025-02-10 Runjin Chen , Gabriel Jacob Perin , Xuxi Chen , Xilun Chen , Yan Han , Nina S. T. Hirata , Junyuan Hong , Bhavya Kailkhura

A trustworthy real-world prediction system should produce well-calibrated confidence scores; that is, its confidence in an answer should be indicative of the likelihood that the answer is correct, enabling deferral to an expert in cases of…

Computation and Language · Computer Science 2023-10-25 Katherine Tian , Eric Mitchell , Allan Zhou , Archit Sharma , Rafael Rafailov , Huaxiu Yao , Chelsea Finn , Christopher D. Manning

It is well-known that neural networks can unintentionally memorize their training examples, causing privacy concerns. However, auditing memorization in large non-auto-regressive automatic speech recognition (ASR) models has been challenging…

Machine Learning · Computer Science 2023-10-19 Lun Wang , Om Thakkar , Rajiv Mathews

Memorization presents a challenge for several constrained Natural Language Generation (NLG) tasks such as Neural Machine Translation (NMT), wherein the proclivity of neural models to memorize noisy and atypical samples reacts adversely with…

Computation and Language · Computer Science 2022-10-25 Vikas Raunak , Arul Menezes

Recent alignment studies commonly remove introductory boilerplate phrases from supervised fine-tuning (SFT) datasets. This work challenges that assumption. We hypothesize that safety- and reasoning-oriented prefix sentences serve as…

Computation and Language · Computer Science 2026-01-06 Raj Vardhan Tomar , Preslav Nakov , Yuxia Wang

Overconfidence has been shown to impair generalization and calibration of a neural network. Previous studies remedy this issue by adding a regularization term to a loss function, preventing a model from making a peaked distribution. Label…

Machine Learning · Computer Science 2022-10-26 Dongkyu Lee , Ka Chun Cheung , Nevin L. Zhang

Fine-tuning adapts a pretrained machine learning model to a small, sensitive dataset, but this process risks memorizing individual new data points, making the model vulnerable to adversaries who seek to extract sensitive information. In…

Machine Learning · Computer Science 2026-05-21 Hoang Tran , Jorge Ramirez , Jiayi Wang , Alberto Bocchinfuso , Christopher Stanley , M. Paul Laiu

Large pre-trained language models are successfully being used in a variety of tasks, across many languages. With this ever-increasing usage, the risk of harmful side effects also rises, for example by reproducing and reinforcing…

Computation and Language · Computer Science 2022-09-19 Pieter Delobelle , Bettina Berendt