Search Results - "Stuhlmüller, Andreas"

Learning physical parameters from dynamic scenes by Ullman, Tomer D., Stuhlmüller, Andreas, Goodman, Noah D., Tenenbaum, Joshua B.

“…•A probabilistic programming framework for learning physics at different levels.•Rational approximation models to ideal-observer physics…”
Get full text

Knowledge and Implicature: Modeling Language Understanding as Social Cognition by Goodman, Noah D., Stuhlmüller, Andreas

“…Is language understanding a special case of social cognition? To help evaluate this view, we can formalize it as the rational speech‐act theory: Listeners…”
Get full text

Factored Verification: Detecting and Reducing Hallucination in Summaries of Academic Papers by George, Charlie, Stuhlmüller, Andreas

“…Hallucination plagues even frontier LLMs--but how bad is it really for summarizing academic papers? We evaluate Factored Verification, a simple automated…”
Get full text

Iterated Decomposition: Improving Science Q&A by Supervising Reasoning Processes by Reppert, Justin, Rachbach, Ben, George, Charlie, Stebbing, Luke, Byun, Jungwon, Appleton, Maggie, Stuhlmüller, Andreas

“…Language models (LMs) can perform complex reasoning either end-to-end, with hidden latent state, or compositionally, with transparent intermediate state…”
Get full text

RAFT: A Real-World Few-Shot Text Classification Benchmark by Alex, Neel, Lifland, Eli, Tunstall, Lewis, Thakur, Abhishek, Maham, Pegah, Riedel, C. Jess, Hine, Emmie, Ashurst, Carolyn, Sedille, Paul, Carlier, Alexis, Noetel, Michael, Stuhlmüller, Andreas

“…Large pre-trained language models have shown promise for few-shot learning, completing text-based tasks given only a few task-specific examples. Will models…”
Get full text

C3: Lightweight Incrementalized MCMC for Probabilistic Programs using Continuations and Callsite Caching by Ritchie, Daniel, Stuhlmüller, Andreas, Goodman, Noah D

“…Lightweight, source-to-source transformation approaches to implementing MCMC for probabilistic programming languages are popular for their simplicity, support…”
Get full text

Agent-Agnostic Human-in-the-Loop Reinforcement Learning by Abel, David, Salvatier, John, Stuhlmüller, Andreas, Evans, Owain

“…Providing Reinforcement Learning agents with expert advice can dramatically improve various aspects of learning. Prior work has developed teaching protocols…”
Get full text

Evaluating Compositionality in Sentence Embeddings by Dasgupta, Ishita, Guo, Demi, Stuhlmüller, Andreas, Gershman, Samuel J, Goodman, Noah D

“…An important challenge for human-like AI is compositional semantics. Recent research has attempted to address this by using deep neural networks to learn…”
Get full text

A Dynamic Programming Algorithm for Inference in Recursive Probabilistic Programs by Stuhlmüller, Andreas, Goodman, Noah D

“…We describe a dynamic programming algorithm for computing the marginal distribution of discrete probabilistic programs. This algorithm takes a functional…”
Get full text

Coarse-to-Fine Sequential Monte Carlo for Probabilistic Programs by Stuhlmüller, Andreas, Hawkins, Robert X. D, Siddharth, N, Goodman, Noah D

“…Many practical techniques for probabilistic inference require a sequence of distributions that interpolate between a tractable distribution and an intractable…”
Get full text

Inducing Probabilistic Programs by Bayesian Program Merging by Hwang, Irvin, Stuhlmüller, Andreas, Goodman, Noah D

“…This report outlines an approach to learning generative models from data. We express models as probabilistic programs, which allows us to capture abstract…”
Get full text