Results 281 to 290 of about 531,339 (320)

The factors of Slavic and Turkic languages typological complementary in the light of systemic linguistics and ethnology

Philological Sciences Scientific Essays of Higher Education, 2019
Relying on the achievements of system linguistics and ethnology, the authors of the article, referring to the history of Nostratic languages, discuss the reasons for the typological divergence of originally related languages and the typological stability
U. Bakhtikireeva, O. Valentinova, Michael A. Rybakov   +2 more
semanticscholar   +1 more source

Linguistic Divergence and the Collapse of Preclassic Civilization in Southern Mesoamerica

American Antiquity, 1987
Based on published lexicostatistical dates, two intervals in the prehistory of southern Mesoamerica stand out as fertile periods in terms of the generation of new languages: the Terminal Preclassic/early Early Classic Periods, and the Early Postclassic Period.
Bruce H. Dahlin, Robin Quizar, Andrea Dahlin   +2 more
openaire   +1 more source

Linguistic divergence under contact

2019
Abstract The normal result of language contact is widely assumed to be convergence, as manifested in classic Sprachbünde and caused through metatypy, cognitive economy, shared norms of conversational practice, etc. Yet at the same time there is growing evidence that contact can also produce divergence, originating with Larsen’s idea of ‘neighbour ...
openaire   +1 more source

Specializing Large Language Models to Simulate Survey Response Distributions for Global Populations

North American Chapter of the Association for Computational Linguistics
Large-scale surveys are essential tools for informing social science research and policy, but running surveys is costly and time-intensive. If we could accurately simulate group-level survey results, this would therefore be very valuable to social ...
Yong Cao, Haijiang Liu, Arnav Arora, Isabelle Augenstein, Paul Rottger, Daniel Hershcovich   +5 more
semanticscholar   +1 more source

BitDistiller: Unleashing the Potential of Sub-4-Bit LLMs via Self-Distillation

Annual Meeting of the Association for Computational Linguistics
The upscaling of Large Language Models (LLMs) has yielded impressive advances in natural language processing, yet it also poses significant deployment challenges.
Dayou Du, Yijia Zhang, Shijie Cao, Jiaqi Guo, Ting Cao, Xiaowen Chu, Ningyi Xu   +6 more
semanticscholar   +1 more source

DIALECTBENCH: A NLP Benchmark for Dialects, Varieties, and Closely-Related Languages

Annual Meeting of the Association for Computational Linguistics
Language technologies should be judged on their usefulness in real-world use cases. An often overlooked aspect in natural language processing (NLP) research and evaluation is language variation in the form of non-standard dialects or language varieties ...
FAHIM FAISAL, Orevaoghene Ahia, Aarohi Srivastava, Kabir Ahuja, David Chiang, Yulia Tsvetkov, Antonios Anastasopoulos   +6 more
semanticscholar   +1 more source

Enough Coin Flips Can Make LLMs Act Bayesian

Annual Meeting of the Association for Computational Linguistics
Large language models (LLMs) exhibit the ability to generalize given few-shot examples in their input prompt, an emergent capability known as in-context learning (ICL).
Ritwik Gupta, Rodolfo Corona, Jiaxin Ge, Eric Wang, Dan Klein, Trevor Darrell, David M. Chan   +6 more
semanticscholar   +1 more source

AdaMergeX: Cross-Lingual Transfer with Large Language Models via Adaptive Adapter Merging

North American Chapter of the Association for Computational Linguistics
As an effective alternative to the direct fine-tuning on target tasks in specific languages, cross-lingual transfer addresses the challenges of limited training data by decoupling ''task ability'' and ''language ability'' by fine-tuning on the target ...
Yiran Zhao, Wenxuan Zhang, Huiming Wang, Kenji Kawaguchi, Lidong Bing   +4 more
semanticscholar   +1 more source

AdaCAD: Adaptively Decoding to Balance Conflicts between Contextual and Parametric Knowledge

North American Chapter of the Association for Computational Linguistics
Knowledge conflict arises from discrepancies between information in the context of a large language model (LLM) and the knowledge stored in its parameters.
Han Wang, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal   +3 more
semanticscholar   +1 more source

Revisiting Weak-to-Strong Generalization in Theory and Practice: Reverse KL vs. Forward KL

Annual Meeting of the Association for Computational Linguistics
As large language models advance toward superhuman performance, ensuring their alignment with human values and abilities grows increasingly complex. Weak-to-strong generalization offers a promising approach by leveraging predictions from weaker models to
Wei Yao, Wenkai Yang, Ziqiao Wang, Yankai Lin, Yong Liu   +4 more
semanticscholar   +1 more source