method · first seen 5 Oct 2026
continued pretraining
Also: domain-adaptive pretraining
Further pretraining of an already pretrained model on domain-specific corpora to adapt it to a target domain.
Mentions per week
PapersCodeDiscussion
Rank 64 on 9 Nov 2026 · 6 mentions in 28 days · 5 independent groups · papers · first seen 5 weeks ago · acceleration +3.33
Earliest sources
- Domain adaptation of Russian ModernBERT for long legal documentsPapers · 5 Oct 2026 · I. Litvak, D. Gvozdetsky, F. Lashkin, V. Kirova, S. Lagutin, V. Volf, T. Maksiya…
- Beyond the greenwash: do LLMs' ecologically sound decisions hold up under pressure?Papers · 12 Oct 2026 · Trent R. Northen, Mingxun Wang
- HexaBee-4B: Six-Language Translation with Multilingual Answering and Bidirectional SupervisionPapers · 12 Oct 2026 · Iman Barati, Mohammadmostafa Rostamkhani, Mohammad Badzohreh
- Teach reality, not ethics: a drop-in continued-pretraining corpus installs coupled human-biosphere-AI welfare dispositions across model familiesPapers · 27 Oct 2026 · Trent R. Northen, Mingxun Wang
- Assessing Language Models for Scientific Applications: Inertial Confinement Fusion as a Use CasePapers · 27 Oct 2026 · P. B. Radha, D. D. Meyerhofer, M. Bhattarai, B. Cheng, L. Diaz, C. L. Rousculp,…
- Saidzoda-Tajik-LFM-2.6B: Adapting a 2.6B Hybrid Model to Tajik Tool Calling, and What Continued Pretraining Takes AwayPapers · 30 Oct 2026 · Tohir Saidzoda, Nekruz Maxudov, Bedil Karimov, Abubakr Abdullozoda, Amin Jalilov…
Latest sources
- Adapting Qwen3 to Tajik from 0.6B to 8B: Lower Text Loss Did Not Mean Better UnderstandingPapers · 4 Nov 2026 · Tohir Saidzoda, Madina Abdusalomzoda, Nasimjon Rajabov, Akhrorkhon Ikromov, Komr…
- Learning from Unlabeled Photographs for General Oral Image AnalysisPapers · 3 Nov 2026 · Zhong-Shi Zhang, Zi-Li Yu, Xiao Xiao, Huan Liu, Gui-Song Xia, Cui Huang, Meng-Xu…
- BanglishDepYT: Depression Detection in Native Bengali Social Media with Emotion-Preserving Preprocessing and Domain-Adaptive PretrainingPapers · 1 Nov 2026 · Ayon Roy, Sadman Sharif, Syed Shams Islam
- Saidzoda-Tajik-LFM-2.6B: Adapting a 2.6B Hybrid Model to Tajik Tool Calling, and What Continued Pretraining Takes AwayPapers · 30 Oct 2026 · Tohir Saidzoda, Nekruz Maxudov, Bedil Karimov, Abubakr Abdullozoda, Amin Jalilov…
- Assessing Language Models for Scientific Applications: Inertial Confinement Fusion as a Use CasePapers · 27 Oct 2026 · P. B. Radha, D. D. Meyerhofer, M. Bhattarai, B. Cheng, L. Diaz, C. L. Rousculp,…
- Teach reality, not ethics: a drop-in continued-pretraining corpus installs coupled human-biosphere-AI welfare dispositions across model familiesPapers · 27 Oct 2026 · Trent R. Northen, Mingxun Wang
- HexaBee-4B: Six-Language Translation with Multilingual Answering and Bidirectional SupervisionPapers · 12 Oct 2026 · Iman Barati, Mohammadmostafa Rostamkhani, Mohammad Badzohreh
- Beyond the greenwash: do LLMs' ecologically sound decisions hold up under pressure?Papers · 12 Oct 2026 · Trent R. Northen, Mingxun Wang
- Domain adaptation of Russian ModernBERT for long legal documentsPapers · 5 Oct 2026 · I. Litvak, D. Gvozdetsky, F. Lashkin, V. Kirova, S. Lagutin, V. Volf, T. Maksiya…