collaborators

6 papers

cs.AI2026

Position: AI Evaluations Should be Grounded on a Theory of Capability

Nathanael Jo, Ashia Wilson

Evaluations of generative models are now ubiquitous, and their outcomes critically shape public and scientific expectations of AI's capabilities. Yet skepticism about their reliabi…

cs.LG2026

Evaluation without Generation: Non-Generative Assessment of Harmful Model Specialization with Applications to CSAM

Vinith M. Suriyakumar, Ayush Sekhari, Lena Stempfle +5

Auditing the fine-tunes of open-weight generative models for harmful specialization has become a new governance challenge for model hosting platforms. The standard toolkit, generat…

cs.LG2026

Online Reasoning Calibration: Test-Time Training Enables Generalizable Conformal LLM Reasoning

Cai Zhou, Zekai Wang, Menghua Wu +6

While test-time scaling has enabled large language models to solve highly difficult tasks, state-of-the-art results come at exorbitant compute costs. These inefficiencies can be at…

cs.LG2026

Efficient and accurate steering of Large Language Models through attention-guided feature learning

Parmida Davarmanesh, Ashia Wilson, Adityanarayanan Radhakrishnan

Steering, or direct manipulation of internal activations to guide LLM responses toward specific semantic concepts, is emerging as a promising avenue for both understanding how sema…

cs.CL2025

UCD: Unlearning in LLMs via Contrastive Decoding

Vinith M. Suriyakumar, Ayush Sekhari, Ashia Wilson

Machine unlearning aims to remove specific information, e.g. sensitive or undesirable content, from large language models (LLMs) while preserving overall performance. We propose an…

cs.LG2025

Layered Unlearning for Adversarial Relearning

Timothy Qian, Vinith Suriyakumar, Ashia Wilson +1

Our goal is to understand how post-training methods, such as fine-tuning, alignment, and unlearning, modify language model behavior and representations. We are particularly interes…