collaborators

5 papers

cs.AI2026

-Monitor: Dynamic Safety Monitoring for Diffusion LLMs via Hesitation-Aware Routing

Aoxi Liu, Yupeng Chen, James Oldfield +5

Despite the emergence of diffusion large language models (D-LLMs) as an alternative to autoregressive large language models (AR-LLMs), safety monitoring for D-LLMs remains largely…

cs.LG2026

PolySAE: Modeling Feature Interactions in Sparse Autoencoders via Polynomial Decoding

Panagiotis Koromilas, Andreas D. Demou, James Oldfield +2

Sparse autoencoders (SAEs) interpret neural network representations by decomposing activations into sparse combinations of dictionary atoms. However, SAEs assume features combine a…

cs.LG2026

fmxcoders: Factorized Masked Crosscoders for Cross-Layer Feature Discovery

Andreas D. Demou, Panagiotis Koromilas, James Oldfield +2

Many features in pretrained Transformers span multiple layers: they emerge through stages of inference, persist in the residual stream, or are built jointly by parallel MLPs. Cross…

cs.LG2026

Beyond Linear Probes: Dynamic Safety Monitoring for Language Models

James Oldfield, Philip Torr, Ioannis Patras +2

Monitoring large language models' (LLMs) activations is an effective way to detect harmful requests before they lead to unsafe outputs. However, traditional safety monitors often r…

cs.LG2026

Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders

James Oldfield, Shawn Im, Sharon Li +3

Multilayer perceptrons (MLPs) are an integral part of large language models, yet their dense representations render them difficult to understand, edit, and steer. Recent methods le…