Research map / Probabilistic and scientific modeling / Variational and generative sampling
Divide, Conquer, and Combine: a New Inference Strategy for Probabilistic Programs with Stochastic Support
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Divide stochastic-support inference into fixed-support sub-models, infer locally, and combine estimates | Probabilistic programs with stochastic support · Function induction, test data | LPPD: -32.693 $\pm$ 8.51 -> -28.560 $\pm$ 0.41 | Helped |
| Divide inference by execution path and combine local estimates | DCC versus IS · Function induction, test data | LPPD: -73.180 $\pm$ 1.08 -> -28.560 $\pm$ 0.41 | Helped |
| Divide inference by execution path and combine local estimates | DCC versus IPMCMC · Function induction, test data | LPPD: -70.580 $\pm$ 4.71 -> -28.560 $\pm$ 0.41 | Helped |
| Divide inference by execution path and combine local estimates | DCC versus PG · Function induction, test data | LPPD: -200.824 $\pm$ 126.63 -> -28.560 $\pm$ 0.41 | Helped |
| Use DCC instead of RMH for inference over stochastic-support models | GMM with misspecified prior · GMM with misspecified prior | posterior probability estimate for K=5: 0 -> around 0.9976 | Helped |
Closest papers by meaning
- Sound Abstraction and Decomposition of Probabilistic Programs (ICML 2018 · similarity 0.91)
- Automatic Reparameterisation of Probabilistic Programs (ICML 2020 · similarity 0.89)
- BIRD: A Trustworthy Bayesian Inference Framework for Large Language Models (ICLR 2025 · similarity 0.85)
- OptScale: Probabilistic Optimality for Inference-time Scaling (AAAI 2026 · similarity 0.85)
- Rollout Roulette: A Probabilistic Inference Approach to Inference-Time Scaling of LLMs using Particle-Based Monte Carlo Methods (NeurIPS 2025 · similarity 0.85)
- A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.