NEURIPS / EMNLP / LOG
2026
Self-play. Retrieval.
Steering. Evaluation.
Self-play

Learning without
curated answers

Retrieval

Inside text + image
attention heads

Steering

Composing multiple
behaviours

Evaluation

Beyond headline
benchmark scores

Graph learning

Interfaces for relational
generalisation

PREPRINTS
SPSD / FACE-Eval / VLM frequency filtering