collaborators

5 papers

cs.CL2026

Annotations Mitigate Post-Training Mode Collapse

Jacob Mitchell Springer, Madhu Advani, Lukas Aichberger +7

Post-training (via supervised fine-tuning) improves instruction-following, but often induces semantic mode collapse by biasing models toward low-entropy fine-tuning data at the exp…

cs.LG2026

Local Mechanisms of Compositional Generalization in Conditional Diffusion

Arwen Bradley

Conditional diffusion models appear capable of compositional generalization, i.e., generating convincing samples for out-of-distribution combinations of conditioners, but the mecha…

cs.CL2025

Trained on Tokens, Calibrated on Concepts: The Emergence of Semantic Calibration in LLMs

Preetum Nakkiran, Arwen Bradley, Adam Goliński +3

Large Language Models (LLMs) often lack meaningful confidence estimates for their outputs. While base LLMs are known to exhibit next-token calibration, it remains unclear whether t…

cs.LG2025

To Infinity and Beyond: Tool-Use Unlocks Length Generalization in State Space Models

Eran Malach, Omid Saremi, Sinead Williamson +5

State Space Models (SSMs) have become the leading alternative to Transformers for sequence modeling. Their primary advantage is efficiency in long-context and long-form generation,…

cs.AI2025

Trace Length is a Simple Uncertainty Signal in Reasoning Models

Siddartha Devic, Charlotte Peale, Arwen Bradley +3

Uncertainty quantification for LLMs is a key research direction towards addressing hallucination and other issues that limit their reliable deployment. In this work, we show that r…