國立虎尾科技大學 |

Challenges and Methods for Alignment of Large Language Models with Human Preferences.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Challenges and Methods for Alignment of Large Language Models with Human Preferences./
Author:	Dang, John Arthur MinhQuan.
Description:	1 online resource (71 pages)
Notes:	Source: Masters Abstracts International, Volume: 85-06.
Contained By:	Masters Abstracts International85-06.
Subject:	Computer engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9798381162301

Challenges and Methods for Alignment of Large Language Models with Human Preferences.
Dang, John Arthur MinhQuan.

Challenges and Methods for Alignment of Large Language Models with Human Preferences. - 1 online resource (71 pages)

Source: Masters Abstracts International, Volume: 85-06.

Thesis (M.S.)--University of California, Los Angeles, 2023.

Includes bibliographical references

While recent advances in methods for aligning Large Language Models (LLMs) have enabled incredible advances in capabilities of widely deployed AI systems, significant challenges still remain in instilling import human values in AI systems including, but not limited to helpful- ness, harmlessness, and fairness. We conduct a study on the impact of feedback acquisition protocols on downstream alignment and evaluation of LLMs. We find that different feed- back acquisition protocols can often produce feedback data which is inconsistent and that feedback data used during alignment biases evaluation of LLMs depending on the feedback acquisition protocol during evaluation. Finally, we introduce Group Preference Optimiza- tion (GPO), a novel method for few-shot aligning LLMs to preferences from groups. We experimentally validate GPO, showing that our method enables efficient alignment of LLM responses to preferences of various demographic groups, given a small amount of preference data from that group.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798381162301Subjects--Topical Terms:

569006
Computer engineering.
Subjects--Index Terms:

Downstream alignmentIndex Terms--Genre/Form:

554714
Electronic books.

Challenges and Methods for Alignment of Large Language Models with Human Preferences.
LDR:02396ntm a22003977 4500 001 1146246
005 20240812064351.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798381162301
035 $a (MiAaPQ)AAI30815273
035 $a AAI30815273
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Dang, John Arthur MinhQuan. $3 1471604
245 1 0 $a Challenges and Methods for Alignment of Large Language Models with Human Preferences.
264 0 $c 2023
300 $a 1 online resource (71 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 85-06.
500 $a Advisor: Grover, Aditya.
502 $a Thesis (M.S.)--University of California, Los Angeles, 2023.
504 $a Includes bibliographical references
520 $a While recent advances in methods for aligning Large Language Models (LLMs) have enabled incredible advances in capabilities of widely deployed AI systems, significant challenges still remain in instilling import human values in AI systems including, but not limited to helpful- ness, harmlessness, and fairness. We conduct a study on the impact of feedback acquisition protocols on downstream alignment and evaluation of LLMs. We find that different feed- back acquisition protocols can often produce feedback data which is inconsistent and that feedback data used during alignment biases evaluation of LLMs depending on the feedback acquisition protocol during evaluation. Finally, we introduce Group Preference Optimiza- tion (GPO), a novel method for few-shot aligning LLMs to preferences from groups. We experimentally validate GPO, showing that our method enables efficient alignment of LLM responses to preferences of various demographic groups, given a small amount of preference data from that group.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer engineering. $3 569006
650 4 $a Computer science. $3 573171
653 $a Downstream alignment
653 $a Generative AI
653 $a Large Language Models
653 $a Machine learning
653 $a Natural language processing
655 7 $a Electronic books. $2 local $3 554714
690 $a 0800
690 $a 0984
690 $a 0464
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of California, Los Angeles. $b Computer Science 0201. $3 1182286
773 0 $t Masters Abstracts International $g 85-06.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30815273 $z click for full text (PQDT)