Upgrade to Pro — share decks privately, control downloads, hide ads and more …

[WIP] How Do Large Language Models Acquire Fact...

Avatar for Yano Yano
August 23, 2025
0
4

[WIP] How Do Large Language Models Acquire Factual Knowledge During Pretraining?

Avatar for Yano

Yano

August 23, 2025
Tweet

More Decks by Yano

See All by Yano
NLP2025参加報告
Avatar for Yano yano0
0
530
【輪講資料】ReAct: Synergizing Reasoning and Acting in Language Models / Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Avatar for Yano yano0
0
170
【輪講資料】SimLM: Pre-training with Representation Bottleneck for Dense Passage Retrieval
Avatar for Yano yano0
2
320
【輪講資料】From Zero to Hero: On the Limitations of Zero-Shot Language Transfer with Multilingual Transformers
Avatar for Yano yano0
0
84
【輪講資料】Zero-shot Cross-lingual Semantic Parsing
Avatar for Yano yano0
0
120

Featured

See All Featured
Facilitating Awesome Meetings
Avatar for Lara Hogan lara
55
6.5k
4 Signs Your Business is Dying
Avatar for Josh Pigford shpigford
184
22k
The Cost Of JavaScript in 2023
Avatar for Addy Osmani addyosmani
53
8.8k
Docker and Python
Avatar for Tania Allard trallard
45
3.5k
How To Stay Up To Date on Web Technology
Avatar for Chris Coyier chriscoyier
790
250k
Mobile First: as difficult as doing things right
Avatar for Swwweet swwweet
223
9.9k
[Rails World 2023 - Day 1 Closing Keynote] - The Magic of Rails
Avatar for Eileen M. Uchitelle eileencodes
36
2.5k
The Art of Programming - Codeland 2020
Avatar for Erika Heidi erikaheidi
54
13k
How to train your dragon (web standard)
Avatar for Monica Dinculescu notwaldorf
96
6.2k
Creating an realtime collaboration tool: Agile Flush - .NET Oxford
Avatar for Marc Duiker marcduiker
31
2.2k
The Art of Delivering Value - GDevCon NA Keynote
Avatar for David Neal reverentgeek
15
1.6k
Building Applications with DynamoDB
Avatar for Matt Wood mza
96
6.6k

Transcript

  1. How Do Large Language Models Acquire Factual Knowledge During Pretraining?

    ࡫໺ݚڀࣨɹD1 ໼໺ઍߛ Hoyeon Chang, Jinho Park, Seonghyeon Ye, Sohee Yang, Youngkyung Seo, Du-Seong Chang, Minjoon Seo NeurIPS 2025