sequential-monte-carlo

Tag

Cards List
#sequential-monte-carlo

Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding

arXiv cs.CL · 4d ago Cached

Chopthin-Consensus Power Sampling (CCPS) is a diversity-preserving method for LLM decoding that improves reasoning accuracy without post-training by preserving distinct reasoning paths and using semantic-majority selection, as demonstrated in benchmarks.

0 favorites 0 likes
#sequential-monte-carlo

Recipes for Steering and Scaling LLMs via Sampling

arXiv cs.CL · 2026-08-28 Cached

The paper presents a theoretical framework for steering and scaling large language models via sampling algorithms, such as Sequential Monte Carlo and Replica Exchange, to improve generation quality without external supervision.

0 favorites 0 likes
#sequential-monte-carlo

LatticeBridge: Rare-Event Sequential Inference for Faithful Structured Sequence Synthesis

arXiv cs.CL · 2026-06-11 Cached

LatticeBridge proposes a twisted sequential Monte Carlo decoder for structured sequence generation that improves constraint satisfaction by treating the problem as rare-event inference, outperforming greedy and beam baselines on CommonGen, E2E NLG, and WikiBio.

0 favorites 0 likes
#sequential-monte-carlo

Agentic Monte Carlo: Simulating Reinforcement Learning for Black-Box Agents

arXiv cs.LG · 2026-06-05 Cached

Introduces Agentic Monte Carlo (AMC), a method to perform reinforcement learning-style optimization of black-box LLM agents using Sequential Monte Carlo, without requiring access to model parameters.

0 favorites 0 likes
#sequential-monte-carlo

SMCEvolve: Principled Scientific Discovery via Sequential Monte Carlo Evolution

arXiv cs.AI · 2026-05-18 Cached

SMCEvolve introduces a principled framework for LLM-driven program evolution by reformulating it as sampling from a reward-tilted distribution using Sequential Monte Carlo. It provides convergence guarantees and outperforms existing methods across multiple scientific discovery benchmarks.

0 favorites 0 likes
#sequential-monte-carlo

Faster LLM Inference via Sequential Monte Carlo

arXiv cs.CL · 2026-04-20 Cached

This paper proposes Sequential Monte Carlo Speculative Decoding (SMC-SD), a method that accelerates LLM inference by replacing token-level rejection in speculative decoding with importance-weighted resampling over draft particles, achieving 2.36× speedup over standard speculative decoding and 5.2× over autoregressive decoding while maintaining 3% accuracy loss.

0 favorites 0 likes
← Back to home

Submit Feedback