notesum.ai

Published at October 28

Flaming-hot Initiation with Regular Execution Sampling for Large Language Models

cs.AI

q-bio.QM

Released Date: October 28, 2024

Authors: Weizhe Chen¹, Zhicheng Zhang², Guanlin Liu³, Renjie Zheng³, Wenlei Shi³, Chen Dun³, Zheng Wu³, Xing Jin³, Lin Yan³

Aff.: ¹University of Southern California; ²Carnegie Mellon University; ³ByteDance

Arxiv: https://arxiv.org/abs/2410.21236v1

Refer to caption

		Regular		FIRE
	Model	Pass%	#EA	Pass%	#EA
	DeepSeek	97.57	2.26	98.71	2.76
GSM8K	Gemma-2	86.81	3.87	87.57	4.01
	Qwen2	95.90	2.58	98.25	3.17
	Qwen2-RL	96.90	2.63	97.90	3.26
	DeepSeek	76.16	5.63	78.16	7.89
MATH	Gemma-2	49.20	9.24	51.48	10.39
	Qwen2	76.60	7.44	79.08	9.03
	Qwen2.5-72B	79.30	2.39	80.40	2.60