paper-with-me

Papers

Synthesizing Post-Training Data for LLMs through Multi-Agent Simulation

2024-10-18 · Shuo Tang, Xianghe Pang, Zexi Liu, Bohan Tang, Rui Ye, Xiaowen Dong, Yanfeng Wang, Siheng Chen

Post-training is essential for enabling large language models (LLMs) to follow human instructions. Inspired by the recent success of using LLMs to simulate human society, we leverage multi-agent simulation to automatically generate diverse text-based scenarios, capturing a wide range of real-world human needs. We propose MATRIX, a multi-agent simulator that creates realistic and scalable scenarios. Leveraging these outputs, we introduce a novel scenario-driven instruction generator MATRIX-Gen for controllable and highly realistic data synthesis. Extensive experiments demonstrate that our framework effectively generates both general and domain-specific data. Notably, on AlpacaEval 2 and Arena-Hard benchmarks, Llama-3-8B-Base, post-trained on datasets synthesized by MATRIX-Gen with just 20K instruction-response pairs, outperforms Meta's Llama-3-8B-Instruct model, which was trained on over 10M pairs; see our project at https://github.com/ShuoTang123/MATRIX-Gen.

📄 PDF Abstract BibTeX arXiv:2410.14251

Code (1)

shuotang123/matrix-gen 공식 구현

Similar Papers 제목 키워드 기반

ReGenesis: LLMs can Grow into Reasoning Generalists via Self-Improvement

2024-10-03 · Xiangyu Peng, Congying Xia, Xinyi Yang, Caiming Xiong 외

Post-training Large Language Models (LLMs) with explicit reasoning trajectories can enhance their reasoning abilities. However, acquiring such high-quality trajectory data typically demands meticulous supervision from hu…

PuzzleClone: A DSL-Powered Framework for Synthesizing Verifiable Data

2025-08-21 · Kai Xiong, Yanwei Huang, Rongjunchen Zhang, Kun Chen 외 arxiv

High-quality mathematical and logical datasets with verifiable answers are essential for strengthening the reasoning capabilities of large language models (LLMs). While recent data augmentation techniques have facilitate…

Data Augmentation

A Survey on Training-free Alignment of Large Language Models

2025-08-12 · Birong Pan, Yongqi Li, Weiyu Zhang, Wenpeng Lu 외 arxiv

The alignment of large language models (LLMs) aims to ensure their outputs adhere to human values, ethical standards, and legal norms. Traditional alignment methods often rely on resource-intensive fine-tuning (FT), whic…

SAND-Math: Using LLMs to Generate Novel, Difficult and Useful Mathematics Questions and Answers

2025-07-28 · Chaitanya Manem, Pratik Prabhanjan Brahma, Prakamya Mishra, Zicheng Liu 외 arxiv

The demand for Large Language Models (LLMs) at multiple scales, capable of sophisticated and sound mathematical reasoning, continues to grow. However, the development of performant mathematical LLMs is often bottlenecked…

Mathematical Reasoning

ProSec: Fortifying Code LLMs with Proactive Security Alignment

2024-11-19 · Xiangzhe Xu, Zian Su, Jinyao Guo, Kaiyuan Zhang 외

While recent code-specific large language models (LLMs) have greatly enhanced their code generation capabilities, the safety of these models remains under-explored, posing potential risks as insecure code generated by th…

Code Generation