English
Related papers

Related papers: Beyond Preserved Accuracy: Evaluating Loyalty and …

200 papers

Pre-trained universal feature extractors, such as BERT for natural language processing and VGG for computer vision, have become effective methods for improving deep learning models without requiring more labeled data. While effective,…

Computation and Language · Computer Science 2020-05-18 Mitchell A. Gordon , Kevin Duh , Nicholas Andrews

The current state-of-the-art defense methods against adversarial examples typically focus on improving either empirical or certified robustness. Among them, adversarially trained (AT) models produce empirical state-of-the-art defense…

Machine Learning · Computer Science 2022-08-02 Jay Nandy , Sudipan Saha , Wynne Hsu , Mong Li Lee , Xiao Xiang Zhu

Multimodal Large Language Models (MLLM) are increasingly deployed in domains where both reliability and efficiency are critical. However, current models remain overconfident, producing highly certain but incorrect answers. At the same time,…

Computer Vision and Pattern Recognition · Computer Science 2026-02-17 Paul Jonas Kurz , Tobias Jan Wieczorek , Mohamed A. Abdelsalam , Rahaf Aljundi , Marcus Rohrbach

Transfer learning under limited data is a challenging setting, where models must adapt to new tasks with minimal supervision. Prior work has primarily focused on improving absolute accuracy in transfer learning. However, empirical evidence…

Quantum Physics · Physics 2026-05-12 Li-An Lo , Li-Yi Hsu , Hsien-Yi Hsieh

Language model pretraining has led to significant performance gains but careful comparison between different approaches is challenging. Training is computationally expensive, often done on private datasets of different sizes, and, as we…

Computation and Language · Computer Science 2019-07-29 Yinhan Liu , Myle Ott , Naman Goyal , Jingfei Du , Mandar Joshi , Danqi Chen , Omer Levy , Mike Lewis , Luke Zettlemoyer , Veselin Stoyanov

In deep learning applications, robustness measures the ability of neural models that handle slight changes in input data, which could lead to potential safety hazards, especially in safety-critical applications. Pre-deployment assessment of…

Software Engineering · Computer Science 2024-04-26 Wenchuan Mu , Kwan Hui Lim

Leveraging shared learning through Massively Multilingual Models, state-of-the-art machine translation models are often able to adapt to the paucity of data for low-resource languages. However, this performance comes at the cost of…

Computation and Language · Computer Science 2022-11-10 Harshita Diddee , Sandipan Dandapat , Monojit Choudhury , Tanuja Ganu , Kalika Bali

Recent developments in natural language representations have been accompanied by large and expensive models that leverage vast amounts of general-domain text through self-supervised pre-training. Due to the cost of applying such models to…

Computation and Language · Computer Science 2019-09-27 Iulia Turc , Ming-Wei Chang , Kenton Lee , Kristina Toutanova

Large language models used for clinical abstraction are sensitive to prompt wording, yet most work treats prompts as fixed and studies uncertainty in isolation. We argue these should be treated jointly. Across two clinical tasks (MedAlign…

Computation and Language · Computer Science 2026-02-02 Arinbjörn Kolbeinsson , Daniel Timbie , Sajjan Narsinghani , Sanjay Hariharan

We demonstrate that learning procedures that rely on aggregated labels, e.g., label information distilled from noisy responses, enjoy robustness properties impossible without data cleaning. This robustness appears in several ways. In the…

Machine Learning · Statistics 2026-05-26 Chen Cheng , John Duchi

This paper describes a new information-theoretic policy evaluation technique for reinforcement learning. This technique converts any compression or density model into a corresponding estimate of value. Under appropriate stationarity and…

Artificial Intelligence · Computer Science 2014-11-20 Joel Veness , Marc G. Bellemare , Marcus Hutter , Alvin Chua , Guillaume Desjardins

To combat adversarial spelling mistakes, we propose placing a word recognition model in front of the downstream classifier. Our word recognition models build upon the RNN semi-character architecture, introducing several new backoff…

Computation and Language · Computer Science 2019-08-30 Danish Pruthi , Bhuwan Dhingra , Zachary C. Lipton

Memorization in large language models (LLMs) makes them vulnerable to data extraction attacks. While pre-training memorization has been extensively studied, fewer works have explored its impact in fine-tuning, particularly for LoRA…

Machine Learning · Computer Science 2025-06-27 Fei Wang , Baochun Li

The development of deep learning algorithms has extensively empowered humanity's task automatization capacity. However, the huge improvement in the performance of these models is highly correlated with their increasing level of complexity,…

Computer Vision and Pattern Recognition · Computer Science 2024-01-19 Eduarda Caldeira , Pedro C. Neto , Marco Huber , Naser Damer , Ana F. Sequeira

Recent findings suggest that diffusion models significantly enhance empirical adversarial robustness. While some intuitive explanations have been proposed, the precise mechanisms underlying these improvements remain unclear. In this work,…

Machine Learning · Computer Science 2025-05-30 Liu Yuezhang , Xue-Xin Wei

We investigate a new linguistic generalization in pre-trained language models (taking BERT (Devlin et al., 2019) as a case study). We focus on degree modifiers (expressions like slightly, very, rather, extremely) and test the hypothesis…

Computation and Language · Computer Science 2023-10-23 Lisa Bylinina , Alexey Tikhonov , Ekaterina Garmash

We conceptualize the process of understanding as information compression, and propose a method for ranking large language models (LLMs) based on lossless data compression. We demonstrate the equivalence of compression length under…

Artificial Intelligence · Computer Science 2024-06-21 Peijia Guo , Ziguang Li , Haibo Hu , Chao Huang , Ming Li , Rui Zhang

Deep networks are well-known to be fragile to adversarial attacks. We conduct an empirical analysis of deep representations under the state-of-the-art attack method called PGD, and find that the attack causes the internal representation to…

Machine Learning · Computer Science 2019-10-29 Chengzhi Mao , Ziyuan Zhong , Junfeng Yang , Carl Vondrick , Baishakhi Ray

The interpretation of deep learning models is a rapidly growing field, with particular interest in language models. There are various approaches to this task, including training simpler models to replicate neural network predictions and…

Computation and Language · Computer Science 2025-01-13 Sergei Kudriashov , Veronika Zykova , Angelina Stepanova , Yakov Raskind , Eduard Klyshinsky

In this paper, we propose Patience-based Early Exit, a straightforward yet effective inference method that can be used as a plug-and-play technique to simultaneously improve the efficiency and robustness of a pretrained language model…

Computation and Language · Computer Science 2020-10-23 Wangchunshu Zhou , Canwen Xu , Tao Ge , Julian McAuley , Ke Xu , Furu Wei