English
Related papers

Related papers: Eliciting Secret Knowledge from Language Models

200 papers

As language models become more powerful and sophisticated, it is crucial that they remain trustworthy and reliable. There is concerning preliminary evidence that models may attempt to deceive or keep secrets from their operators. To explore…

Machine Learning · Computer Science 2025-05-21 Bartosz Cywiński , Emil Ryd , Senthooran Rajamanoharan , Neel Nanda

Empowering large language models to accurately express confidence in their answers is essential for trustworthy decision-making. Previous confidence elicitation methods, which primarily rely on white-box access to internal model information…

Computation and Language · Computer Science 2024-03-19 Miao Xiong , Zhiyuan Hu , Xinyang Lu , Yifei Li , Jie Fu , Junxian He , Bryan Hooi

Large language models (LLMs) can "lie", which we define as outputting false statements despite "knowing" the truth in a demonstrable sense. LLMs might "lie", for example, when instructed to output misinformation. Here, we develop a simple…

Computation and Language · Computer Science 2023-09-28 Lorenzo Pacchiardi , Alex J. Chan , Sören Mindermann , Ilan Moscovitz , Alexa Y. Pan , Yarin Gal , Owain Evans , Jan Brauner

Capability evaluations are required to understand and regulate AI systems that may be deployed or further developed. Therefore, it is important that evaluations provide an accurate estimation of an AI system's capabilities. However, in…

Artificial Intelligence · Computer Science 2025-07-22 Felix Hofstätter , Teun van der Weij , Jayden Teoh , Rada Djoneva , Henning Bartsch , Francis Rhys Ward

Neural language models are black-boxes--both linguistic patterns and factual knowledge are distributed across billions of opaque parameters. This entangled encoding makes it difficult to reliably inspect, verify, or update specific facts.…

Using AI to create autonomous researchers has the potential to accelerate scientific discovery. A prerequisite for this vision is understanding how well an AI model can identify the underlying structure of a black-box system from its…

Machine Learning · Computer Science 2025-05-26 Jiayi Geng , Howard Chen , Dilip Arumugam , Thomas L. Griffiths

Providing Language Models (LMs) with relevant evidence in the context (either via retrieval or user-provided) can significantly improve their ability to provide better-grounded responses. However, recent studies have found that LMs often…

Computation and Language · Computer Science 2025-05-27 Zhining Liu , Rana Ali Amjad , Ravinarayana Adkathimar , Tianxin Wei , Hanghang Tong

Language Models (LMs) may acquire harmful knowledge, and yet feign ignorance of these topics when under audit. Inspired by the recent discovery of deception-related behaviour patterns in LMs, we aim to train classifiers that detect when a…

Computation and Language · Computer Science 2026-03-24 Dhananjay Ashok , Ruth-Ann Armstrong , Jonathan May

Large language models sometimes produce false or misleading responses. Two approaches to this problem are honesty elicitation -- modifying prompts or weights so that the model answers truthfully -- and lie detection -- classifying whether a…

Machine Learning · Computer Science 2026-03-11 Helena Casademunt , Bartosz Cywiński , Khoi Tran , Arya Jakkli , Samuel Marks , Neel Nanda

Pre-trained Language Models (PLMs) are trained on large amounts of unlabeled data, yet they exhibit remarkable reasoning skills. However, the trustworthiness challenges posed by these black-box models have become increasingly evident in…

Computation and Language · Computer Science 2025-08-26 Yunxiao Zhao , Hao Xu , Zhiqiang Wang , Xiaoli Li , Jiye Liang , Ru Li

To determine the safety of large language models (LLMs), AI developers must be able to assess their dangerous capabilities. But simple prompting strategies often fail to elicit an LLM's full capabilities. One way to elicit capabilities more…

Machine Learning · Computer Science 2024-05-31 Ryan Greenblatt , Fabien Roger , Dmitrii Krasheninnikov , David Krueger

Eliciting information to reduce uncertainty about a latent entity is a critical task in many application domains, e.g., assessing individual student learning outcomes, diagnosing underlying diseases, or learning user preferences. Though…

Computation and Language · Computer Science 2025-07-10 Jimmy Wang , Thomas Zollo , Richard Zemel , Hongseok Namkoong

The remarkable success of pretrained language models has motivated the study of what kinds of knowledge these models learn during pretraining. Reformulating tasks as fill-in-the-blanks problems (e.g., cloze tests) is a natural approach for…

Computation and Language · Computer Science 2020-11-10 Taylor Shin , Yasaman Razeghi , Robert L. Logan , Eric Wallace , Sameer Singh

Large language models (LLMs) acquire a breadth of information across various domains. However, their computational complexity, cost, and lack of transparency often hinder their direct application for predictive tasks where privacy and…

Machine Learning · Computer Science 2025-05-29 Alexander Capstick , Rahul G. Krishnan , Payam Barnaghi

Humans use introspection to evaluate their understanding through private internal states inaccessible to external observers. We investigate whether large language models possess similar privileged knowledge about answer correctness,…

Computation and Language · Computer Science 2026-04-28 Tomer Ashuach , Shai Gretz , Yoav Katz , Yonatan Belinkov , Liat Ein-Dor

Large language models (LLMs) are increasingly expected to function as collaborative partners, engaging in back-and-forth dialogue to solve complex, ambiguous problems. However, current LLMs often falter in real-world settings, defaulting to…

Artificial Intelligence · Computer Science 2025-07-30 Tenghao Huang , Sihao Chen , Muhao Chen , Jonathan May , Longqi Yang , Mengting Wan , Pei Zhou

Large Language Models (LLMs) can generate biased responses. Yet previous direct probing techniques contain either gender mentions or predefined gender stereotypes, which are challenging to comprehensively collect. Hence, we propose an…

Computation and Language · Computer Science 2024-02-20 Xiangjue Dong , Yibo Wang , Philip S. Yu , James Caverlee

Reliably predicting the behavior of language models -- such as whether their outputs are correct or have been adversarially manipulated -- is a fundamentally challenging task. This is often made even more difficult as frontier language…

Machine Learning · Computer Science 2025-12-02 Dylan Sam , Marc Finzi , J. Zico Kolter

Large Language Models (LLMs) have showcased exceptional capabilities in various domains, attracting significant interest from both academia and industry. Despite their impressive performance, the substantial size and computational demands…

Computation and Language · Computer Science 2024-07-03 Chuanpeng Yang , Wang Lu , Yao Zhu , Yidong Wang , Qian Chen , Chenlong Gao , Bingjie Yan , Yiqiang Chen

This work presents a framework for assessing whether large language models (LLMs) encode more factual knowledge in their parameters than what they express in their outputs. While a few studies hint at this possibility, none has clearly…

Computation and Language · Computer Science 2025-08-07 Zorik Gekhman , Eyal Ben David , Hadas Orgad , Eran Ofek , Yonatan Belinkov , Idan Szpektor , Jonathan Herzig , Roi Reichart
‹ Prev 1 2 3 10 Next ›