LAFI 2026
as the organisers put it
The Languages for Inference (LAFI) workshop aims to bring programming-language and machine-learning researchers together to advance all aspects of languages for inference: languages that offer built-in support for expressing probabilistic or differentiable models, and methods for inference and optimization over them, as programs, to ease reasoning, use, and reuse. Topics include but are not limited to: Design of programming languages for probabilistic inference and/or differentiable programming Inference algorithms for probabilistic programming languages, including ones that incorporate a ...
who it's for
probabilistic programming researchers, programming language researchers, machine learning researchers, compiler engineers, systems engineers, formal methods researchers, bayesian statisticians
topics
GPU Gibbs sampling; factor-based intermediate representation; reified scheduling for GPUs; incremental density computation; trace-typed program updates; causal probabilistic programming; interventional and counterfactual queries; sequential Monte Carlo program synthesis; weighted automata exact inference; compilation to Bayesian networks; nominal semantics for automatic differentiation; distributional loop invariants
| title | author | length |
|---|---|---|
| A Design for Massively Parallel Gibbs Sampling on the GPU via Static and Dynamic(…) | Matin Ghavami, Martin C. Rinard, Vikash K. Mansinghka | 11m |
Design enables massively parallel Gibbs sampling on GPUs by combining a factor-based intermediate representation for static graph-coloring to expose… | ||
| Incremental Density Computation for Efficient Programmable Inference | Fabian Zaiser, Vikash K. Mansinghka, Alexander K. Lew | 12m |
Presents a concrete workflow: compute trace types from the probabilistic program, use type-directed change types to represent insertions/removals/replacements… | ||
| A Welcome to Causal Probabilistic Programming | Dario Stein | 59m |
Probabilistic programming is extended to answer interventional and counterfactual queries using concrete tools: Cairo on Pyro (with a Haskell port) for… | ||
| Sequential Monte Carlo Program Synthesis with Refinement Proposals | Maddy Bowers, Mauricio Barba da Costa, Xiaoyan Wang, Joshua B. Tenenbaum, Vikash K. Mansinghka, Armando Solar-Lezama, Alexander K. Lew | 14m |
Extends program synthesis search to include probabilistic programs and evaluates exact marginal likelihood using the PL change named pluck via lazy knowledge… | ||
| Multi-Agent Systems for Traceable Bayesian Workflow | Xianda Sun, Andrew D. Gordon, Hong Ge | 18m |
A multi-agent system assembles prompts and tools to run an end-to-end Bayesian workflow: it generates Stan models, runs prior and simulation-based checks… | ||
| Towards an Equational Calculus of Interventions | Shubh Agrawal, Jialu Bao, Steven Holtzen | 12m |
Presents a typed equational theory that rewrites interventions in probabilistic programs into equivalent do‑free programs by replacing mechanisms with a free… | ||
| Probabilistic Programming Meets Automata Theory: Exact Inference using Weighted Automata | Dominik Geißler, Tobias Winkler | 11m |
Exact inference for a restricted class of discrete probabilistic programs using weighted automata that encode (possibly infinite-support distributions) as… | ||
| Language-Model Probabilistic Programming for Improving Autoformalization via Cycle(…) | Mauricio Barba da Costa, Fabian Zaiser, Katherine Collins, Romir Patel, Timothy J. O'Donnell, Alexander K. Lew, Joshua B. Tenenbaum, Vikash K. Mansinghka, Cameron E. Freer | 12m |
Proposes a GenLM-based decoder that steers a 70B LLM with two real-valued potentials: incremental type-checking via Lean by affixing dummy conclusions at safe… | ||
| Towards Compiling Higher-Order Programs to Bayesian Networks | Claudia Faggian, Gabriele Vanoni | 13m |
Claims an effective compilation from a first‑order probabilistic lambda fragment to Bayesian networks, with a path to higher‑order and recursion, by running a… | ||
| A Word Sampler for Well-Typed Functions | Breandan Considine | 14m |
An exact sampler takes an acyclic finite automaton α∅ and produces functions from a simply-typed first-order language that are uniformly without replacement… | ||
| On Contextual Distances in Randomized Programming: Amplification and Lower Bounds | Houssein Mansour, Raphaëlle Crubillé | 10m |
Proves that contextual distance trivializes (non‑equivalent programs reach distance 1) via a probabilistic construction: build copying k times and empirical… | ||
| Verifying Sampling Algorithms via Distributional Invariants | Daniel Zilken, Tobias Winkler, Kevin Batz, Joost-Pieter Katoen | 09m |
Framework treats probabilistic programs as distribution transformers and proves that the Fast Dice Roller and Fast Loaded Dice Roller produce the intended… | ||
| Typed Abstractions for Causal Probabilistic Programming | Theo Wang, Dario Stein, Eli Bingham, Jack Feser, Ohad Kammar, Michael Lee, Jeremy Yallop | 13m |
A Haskell library reimplements Cairo’s causal features with concrete types, extracting a cosmonad (a reader monad transformer over a probability monad) to… | ||
| Semantic Foundations for Laziness in Discrete Probabilistic Programming | Simon Castellan, Tom Hirschowitz, Hugo Paquet | 13m |
Claims that knowledge compilation and binary decision diagrams follow directly from a denotational semantics that splits sampling into two effects… | ||
| Nominal Semantics for First-class Automatic Differentiation | Jack Czenszak, Alexander K. Lew | 14m |
Formalizes the popular tagged forward-mode AD (used in JAX, Julia, early PyTorch) as a meaning-preserving compiler from a simply typed lambda calculus with… | ||
| nothing matches | ||