English
Related papers

Related papers: White-Box Sensitivity Auditing with Steering Vecto…

200 papers

Activation steering is a popular white-box control technique that modifies model activations to elicit an abstract change in its behavior. It has also become a standard tool in interpretability (e.g., probing truthfulness, or translating…

Artificial Intelligence · Computer Science 2026-05-11 Aayush Mishra , Daniel Khashabi , Anqi Liu

Large Language Models (LLMs) have shown a high capability in answering questions on a diverse range of topics. However, these models sometimes produce biased, ideologized or incorrect responses, limiting their applications if there is no…

Artificial Intelligence · Computer Science 2026-04-08 Xiaotian Zhou , Di Tang , Xiaofeng Wang , Xiaozhong Liu

We introduce AuditBench, an alignment auditing benchmark. AuditBench consists of 56 language models with implanted hidden behaviors. Each model has one of 14 concerning behaviors--such as sycophantic deference, opposition to AI regulation,…

Computation and Language · Computer Science 2026-03-11 Abhay Sheshadri , Aidan Ewart , Kai Fronsdal , Isha Gupta , Samuel R. Bowman , Sara Price , Samuel Marks , Rowan Wang

Trained on a vast amount of data, Large Language models (LLMs) have achieved unprecedented success and generalization in modeling fairly complex textual inputs in the abstract space, making them powerful tools for zero-shot learning. Such…

Computer Vision and Pattern Recognition · Computer Science 2023-05-19 Shervin Ardeshir

Activation-based steering enables Large Language Models (LLMs) to exhibit targeted behaviors by intervening on intermediate activations without retraining. Despite its widespread use, the mechanistic factors that govern when steering…

Computation and Language · Computer Science 2026-03-13 Mehdi Jafari , Hao Xue , Flora Salim

The widespread deployment of Large Language Models (LLMs) has intensified concerns about subtle social biases embedded in their outputs. Existing guardrails often fail when faced with indirect or contextually complex bias-inducing prompts.…

Software Engineering · Computer Science 2025-12-02 Sina Salimian , Gias Uddin , Sumon Biswas , Henry Leung

Latent space steering methods provide a practical approach to controlling large language models by applying steering vectors to intermediate activations, guiding outputs toward desired behaviors while avoiding retraining. Despite their…

Machine Learning · Computer Science 2026-01-13 Shawn Im , Sharon Li

The proliferation of open-source Large Language Models (LLMs) underscores the pressing need for evaluation methods. Existing works primarily rely on external evaluators, focusing on training and prompting strategies. However, a crucial…

Computation and Language · Computer Science 2024-09-30 Hui Huang , Yingqi Qu , Jing Liu , Muyun Yang , Bing Xu , Tiejun Zhao , Wenpeng Lu

Large Language Models (LLMs) inherit explicit and implicit biases from their training datasets. Identifying and mitigating biases in LLMs is crucial to ensure fair outputs, as they can perpetuate harmful stereotypes and misinformation. This…

Machine Learning · Computer Science 2025-11-19 Fatima Kazi , Alex Young , Yash Inani , Setareh Rafatirad

Algorithm audits are powerful tools for studying black-box systems. While very effective in examining technical components, the method stops short of a sociotechnical frame, which would also consider users as an integral and dynamic part of…

Human-Computer Interaction · Computer Science 2023-08-31 Michelle S. Lam , Ayush Pandit , Colin H. Kalicki , Rachit Gupta , Poonam Sahoo , Danaë Metaxa

Many modern software systems are highly configurable, allowing the user to tune them for performance and more. Current performance modeling approaches aim at finding performance-optimal configurations by building performance models in a…

Software Engineering · Computer Science 2021-02-15 Max Weber , Sven Apel , Norbert Siegmund

Ensuring trustworthiness in open-world visual recognition requires models that are interpretable, fair, and robust to distribution shifts. Yet modern vision systems are increasingly deployed as proprietary black-box APIs, exposing only…

Computer Vision and Pattern Recognition · Computer Science 2026-04-16 Simone Carnemolla , Chiara Russo , Simone Palazzo , Quentin Bouniot , Daniela Giordano , Zeynep Akata , Matteo Pennisi , Concetto Spampinato

Large Language Models (LLMs) reproduce social biases, yet prevailing evaluations score models in isolation, obscuring how biases persist across families and releases. We introduce Bias Similarity Measurement (BSM), which treats fairness as…

Machine Learning · Computer Science 2025-09-26 Hyejun Jeong , Shiqing Ma , Amir Houmansadr

The objectives that Large Language Models (LLMs) implicitly optimize remain dangerously opaque, making trustworthy alignment and auditing a grand challenge. While Inverse Reinforcement Learning (IRL) can infer reward functions from…

Machine Learning · Computer Science 2025-10-09 Matthieu Bou , Nyal Patel , Arjun Jagota , Satyapriya Krishna , Sonali Parbhoo

This paper surveys evaluation techniques to enhance the trustworthiness and understanding of Large Language Models (LLMs). As reliance on LLMs grows, ensuring their reliability, fairness, and transparency is crucial. We explore algorithmic…

Computation and Language · Computer Science 2024-06-05 Nik Bear Brown

Vision Language Models (VLMs) have demonstrated remarkable capabilities in various open-vocabulary tasks, yet their zero-shot performance lags behind task-specific fine-tuned models, particularly in complex tasks like Referring Expression…

Computer Vision and Pattern Recognition · Computer Science 2025-03-07 Amaia Cardiel , Eloi Zablocki , Elias Ramzi , Oriane Siméoni , Matthieu Cord

We study secret elicitation: discovering knowledge that an AI possesses but does not explicitly verbalize. As a testbed, we train three families of large language models (LLMs) to possess specific knowledge that they apply downstream but…

Machine Learning · Computer Science 2025-11-03 Bartosz Cywiński , Emil Ryd , Rowan Wang , Senthooran Rajamanoharan , Neel Nanda , Arthur Conmy , Samuel Marks

Consider the problem of testing whether the outputs of a large language model (LLM) system change under an arbitrary intervention, such as an input perturbation or changing the model variant. We cannot simply compare two LLM outputs since…

Computation and Language · Computer Science 2025-06-10 Paulius Rauba , Qiyao Wei , Mihaela van der Schaar

Controlling the behavior of large language models (LLMs) at inference time is essential for aligning outputs with human abilities and safety requirements. \emph{Activation steering} provides a lightweight alternative to prompt engineering…

Artificial Intelligence · Computer Science 2026-01-30 Diaoulé Diallo , Katharina Dworatzyk , Sophie Jentzsch , Peer Schütt , Sabine Theis , Tobias Hecking

Due to the implement of guardrails by developers, Large language models (LLMs) have demonstrated exceptional performance in explicit bias tests. However, bias in LLMs may occur not only explicitly, but also implicitly, much like humans who…

Computation and Language · Computer Science 2025-03-05 Xinru Lin , Luyang Li