TY - GEN
T1 - First Tragedy, then Parse
T2 - 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2024
AU - Saphra, Naomi
AU - Fleisig, Eve
AU - Cho, Kyunghyun
AU - Lopez, Adam
N1 - Publisher Copyright:
© 2024 Association for Computational Linguistics.
PY - 2024
Y1 - 2024
N2 - Many NLP researchers are experiencing an existential crisis triggered by the astonishing success of ChatGPT and other systems based on large language models (LLMs). After such a disruptive change to our understanding of the field, what is left to do? Taking a historical lens, we look for guidance from the first era of LLMs, which began in 2005 with large ngram models for machine translation (MT). We identify durable lessons from the first era, and more importantly, we identify evergreen problems where NLP researchers can continue to make meaningful contributions in areas where LLMs are ascendant. We argue that disparities in scale are transient and researchers can work to reduce them; that data, rather than hardware, is still a bottleneck for many applications; that meaningful realistic evaluation is still an open problem; and that there is still room for speculative approaches.
AB - Many NLP researchers are experiencing an existential crisis triggered by the astonishing success of ChatGPT and other systems based on large language models (LLMs). After such a disruptive change to our understanding of the field, what is left to do? Taking a historical lens, we look for guidance from the first era of LLMs, which began in 2005 with large ngram models for machine translation (MT). We identify durable lessons from the first era, and more importantly, we identify evergreen problems where NLP researchers can continue to make meaningful contributions in areas where LLMs are ascendant. We argue that disparities in scale are transient and researchers can work to reduce them; that data, rather than hardware, is still a bottleneck for many applications; that meaningful realistic evaluation is still an open problem; and that there is still room for speculative approaches.
UR - http://www.scopus.com/inward/record.url?scp=85200205918&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85200205918&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85200205918
T3 - Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2024
SP - 2310
EP - 2326
BT - Long Papers
A2 - Duh, Kevin
A2 - Gomez, Helena
A2 - Bethard, Steven
PB - Association for Computational Linguistics (ACL)
Y2 - 16 June 2024 through 21 June 2024
ER -