← Search

Liam Hebert

2 accepted papers

2026

Efficient, Property-Aligned Fan-Out Retrieval via RL-Amortized Diffusion

ICML 2026poster

Many modern retrieval problems are \emph{set-valued}: given a broad intent, the system must return a \emph{collection} of results that optimizes higher-order properties (e.g., diversity, coverage, complementarity, coherence) while staying grounded to a fixed database. Set-valued objectives are inher…

Cited by 0SourceScholar
2024

Multi-Modal Discussion Transformer: Integrating Text, Images and Graph Transformers to Detect Hate Speech on Social Media

AAAI 2024technical

We present the Multi-Modal Discussion Transformer (mDT), a novel method for detecting hate speech on online social networks such as Reddit discussions. In contrast to traditional comment-only methods, our approach to labelling a comment as hate speech involves a holistic analysis of text and images…