Thomas Hartvigsen

Estimating Semantic Alphabet Size for LLM Uncertainty Quantification

Improving and Accelerating Offline RL in Large Discrete Action Spaces with Structured Policy Initialization

SAINT: Attention-Based Policies for Discrete Combinatorial Action Spaces

Adapting and Explaining Time Series Foundation Models with Concepts

BEDTime: A Unified Benchmark for Automatically Describing Time Series

TS2Code: Enhancing Time Series Understanding via Learning to Code

BraVE: Offline Reinforcement Learning for Discrete Combinatorial Action Spaces

WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs

PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages

KScope: A Framework for Characterizing the Knowledge Status of Language Models

Constrained Discrete Diffusion

Learning under Temporal Label Noise

Composable Interventions for Language Models

BalancEdit: Dynamically Balancing the Generality-Locality Trade-off in Multi-modal Model Editing

Backdoor in Seconds: Unlocking Vulnerabilities in Large Pre-trained Models via Model Editing

Aligning Multimodal Models for Clinical Reasoning using Rule-based Rewards

Offline Reinforcement Learning With Combinatorial Action Spaces

BalancEdit: Dynamically Balancing the Generality-Locality Trade-off in Multi-modal Model Editing