Kilian Q Weinberger

Pre-training Limited Memory Language Models with Internal and External Knowledge

SpeechOp: Inference-Time Task Composition for Generative Speech Processing

Learning from Synthetic Data Improves Multi-hop Reasoning

Adaptive Moments are Surprisingly Effective for Plug-and-Play Diffusion Sampling

$\pi$-CoT: Prolog-Initialized Chain-of-Thought Prompting for Multi-Hop Question-Answering

Benchmark Datasets for Lead-Lag Forecasting on Social Platforms

$Q\sharp$: Provably Optimal Distributional RL for LLM Post-Training

$Q\sharp$: Provably Optimal Distributional RL for LLM Post-Training

On Speeding Up Language Model Evaluation

PhantomWiki: On-Demand Datasets for Reasoning and Retrieval Evaluation

Rethinking LLM Unlearning Objectives: A Gradient Perspective and Go Beyond

Learning 3D Perception from Others' Predictions

On Orchestrating Personalized LLMs

Zero-shot Object-level Out-of-distribution Detection with Context-aware Inpainting

Stop-Think-AutoRegress: Language Modeling with Latent Diffusion Planning

FAVEN: Fast Audio-Visual Embodied Navigation in 3D Environments

Hardware Simulation for Analog Ultrasonic 2D Convolution