SSR: Path-Factorized Self-Correction for LLMs

An Old Trick from 400 BC for Next-Gen AI.

All rights for SSR w/ authors:
SSR: SOCRATIC SELF-REFINE FOR LARGE LANGUAGE MODEL REASONING
Haizhou Shi∗ †12 Ye Liu1 Bo Pang1 Zeyu Leo Liu∗13 Hao Wang2
Silvio Savarese1 Caiming Xiong1 Yingbo Zhou1 Semih Yavuz†1
from
1 Salesforce AI Research
2 Rutgers University
3 The University of Texas at Austin

@salesforce #salesforce

All further ideas on renormalization of AI are my own ideas.

#airesearch
#discoverai
#aiexplained
#scienceexplained
#physics

Видео SSR: Path-Factorized Self-Correction for LLMs канала Discover AI