Παρακολούθηση
Jeremy R Cole
Jeremy R Cole
Google DeepMind
Η διεύθυνση ηλεκτρονικού ταχυδρομείου έχει επαληθευτεί στον τομέα google.com - Αρχική σελίδα
Τίτλος
Παρατίθεται από
Παρατίθεται από
Έτος
Gemini: a family of highly capable multimodal models
G Team, R Anil, S Borgeaud, JB Alayrac, J Yu, R Soricut, J Schalkwyk, ...
arXiv preprint arXiv:2312.11805, 2023
25132023
Time-aware language models as temporal knowledge bases
B Dhingra, JR Cole, JM Eisenschlos, D Gillick, J Eisenstein, WW Cohen
Transactions of the Association for Computational Linguistics 10, 257-273, 2022
2542022
Gecko: Versatile text embeddings distilled from large language models
J Lee, Z Dai, X Ren, B Chen, D Cer, JR Cole, K Hui, M Boratko, ...
arXiv preprint arXiv:2403.20327, 2024
622024
Selectively answering ambiguous questions
JR Cole, MJQ Zhang, D Gillick, JM Eisenschlos, B Dhingra, J Eisenstein
arXiv preprint arXiv:2305.14613, 2023
422023
Is Word Adoption a Grassroots Process? An Analysis of Reddit Communities
JR Cole, M Ghafurian, D Reitter
International Conference on Social Computing, Behavioral-Cultural Modeling …, 2017
162017
Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?
J Lee, A Chen, Z Dai, D Dua, DS Sachan, M Boratko, Y Luan, SMR Arnold, ...
arXiv preprint arXiv:2406.13121, 2024
142024
Surprisal Predicts Code-Switching in Chinese-English Bilingual Text
J Calvillo, L Fang, JR Cole, D Reitter
Proceedings of the 2020 Conference on Empirical Methods in Natural Language …, 2020
142020
Word adoption in online communities
JR Cole, M Ghafurian, D Reitter
IEEE Transactions on Computational Social Systems 6 (1), 178-188, 2019
102019
Not that much power: Linguistic alignment is influenced more by low-level linguistic features rather than social power
Y Xu, JR Cole, D Reitter
Proceedings of the 56th Annual Meeting of the Association for Computational …, 2018
102018
Salient Span Masking for Temporal Understanding
JR Cole, A Chaudhary, B Dhingra, P Talukdar
Proceedings of the 17th Conference of the European Chapter of the …, 2023
92023
Do ever larger octopi still amplify reporting biases? Evidence from judgments of typical colour
F Liu, JM Eisenschlos, JR Cole, N Collier
arXiv preprint arXiv:2209.12786, 2022
6*2022
WinoDict: Probing language models for in-context word acquisition
JM Eisenschlos, JR Cole, F Liu, WW Cohen
arXiv preprint arXiv:2209.12153, 2022
62022
The role of working memory in syntactic sentence realization: A modeling & simulation approach
JR Cole, D Reitter
Cognitive Systems Research 55, 95-106, 2019
62019
The timing of lexical memory retrievals in language production
JR Cole, D Reitter
Proceedings of the 2018 Conference of the North American Chapter of the …, 2018
62018
Linking Memory Activation and Word Adoption in Social Language Use via Rational Analysis
JR Cole, M Ghafurian, D Reitter
Proceedings of the International Conference on Cognitive Modelling, 2017
62017
Examining Working Memory during Sentence Construction with an ACT-R Model of Grammatical Encoding
JR Cole, D Reitter
Proceedings of the International Conference on Cognitive Modelling, 2017
52017
How People Talk about Armed Conflicts: An Analysis of Reddit data
JR Cole, Y Xu, D Reitter
International Conference on Social Computing, Behavioral-Cultural Modeling …, 2016
5*2016
Can long-context language models subsume retrieval, rag, sql, and more?, 2024
J Lee, A Chen, Z Dai, D Dua, DS Sachan, M Boratko, Y Luan, SMR Arnold, ...
URL https://arxiv. org/abs/2406.13121, 0
5
DIFFQG: Generating Questions to Summarize Factual Changes
JR Cole, P Jain, JM Eisenschlos, MJQ Zhang, E Choi, B Dhingra
arXiv preprint arXiv:2303.00242, 2023
32023
NAIL: Lexical Retrieval Indices with Efficient Non-Autoregressive Decoders
LB Soares, D Gillick, JR Cole, T Kwiatkowski
arXiv preprint arXiv:2305.14499, 2023
22023
Δεν είναι δυνατή η εκτέλεση της ενέργειας από το σύστημα αυτή τη στιγμή. Προσπαθήστε ξανά αργότερα.
Άρθρα 1–20