paper-with-me

Papers

Improving Constrained Language Generation via Self-Distilled Twisted Sequential Monte Carlo

2025-07-03 · Sooyeon Kim, Giung Nam, Byoungwoo Park, Juho Lee arxiv

Recent work has framed constrained text generation with autoregressive language models as a probabilistic inference problem. Among these, Zhao et al. (2024) introduced a promising approach based on twisted Sequential Monte Carlo, which incorporates learned twist functions and twist-induced proposals to guide the generation process. However, in constrained generation settings where the target distribution concentrates on outputs that are unlikely under the base model, learning becomes challenging due to sparse and uninformative reward signals. We show that iteratively refining the base model through self-distillation alleviates this issue by making the model progressively more aligned with the target, leading to substantial gains in generation quality.

📄 PDF Abstract BibTeX arXiv:2507.02315

Code (0)

등록된 구현이 없습니다.

Tasks

Text Generation

Similar Papers 제목 키워드 기반

Inductance-Based Force Self-Sensing in Fiber-Reinforced Pneumatic Twisted-and-Coiled Actuators

2026-03-19 · Yunsong Zhang, Tianlin Li, Mingyang Yang, Feitian Zhang arxiv

Fiber-reinforced pneumatic twisted-and-coiled actuators (FR-PTCAs) offer high power density and compliance but their strong hysteresis and lack of intrinsic proprioception limit effective closed-loop control. This paper …

Inference-Time Alignment of Diffusion Models via Trust-Region Iterative Twisted Sequential Monte Carlo

2026-05-24 · Weixin Wang, Yu Yang, Wei Deng, Pan Xu arxiv

We study inference-time alignment for diffusion-based generative models, aiming to steer a base model toward high-reward outputs without updating its weights. Recent Sequential Monte Carlo (SMC)-based steering methods ap…

Text-to-Image GenerationText Generation

Trust-Region Twisted Policy Improvement

2025-04-08 · Joery A. de Vries, Jinke He, Yaniv Oren, Matthijs T. J. Spaan

Monte-Carlo tree search (MCTS) has driven many recent breakthroughs in deep reinforcement learning (RL). However, scaling MCTS to parallel compute has proven challenging in practice which has motivated alternative planne…

Deep Reinforcement LearningReinforcement Learning (RL)

Mind's Mirror: Distilling Self-Evaluation Capability and Comprehensive Thinking from Large Language Models

2023-11-15 · Weize Liu, Guocong Li, Kai Zhang, Bang Du 외

Large language models (LLMs) have achieved remarkable advancements in natural language processing. However, the massive scale and computational demands of these models present formidable challenges when considering their…

Transfer Learning

Impossible Distillation: from Low-Quality Model to High-Quality Dataset & Model for Summarization and Paraphrasing

2023-05-26 · JaeHun Jung, Peter West, Liwei Jiang, Faeze Brahman 외

We present Impossible Distillation, a novel framework for paraphrasing and sentence summarization, that distills a high-quality dataset and model from a low-quality teacher that itself cannot perform these tasks. Unlike …

DiversitymodelParaphrase GenerationSentence+1