arxiv:2609.06970
Jed Cheng PRO
jed351
AI & ML interests
Cantonese used in Hong Kong
Recent Activity
updated a dataset about 18 hours ago
jed351/Traditional-Chinese-Common-Crawl-NOT-Cleaned updated a model 30 days ago
jed351/skyrmion-unetpp-moke published a model 30 days ago
jed351/skyrmion-unetpp-mokeOrganizations
models 23
jed351/skyrmion-unetpp-moke
Image Segmentation • 9.17M • Updated • 36
jed351/qwen3-4B-cpt-v2-530-steps
4B • Updated • 5
jed351/cantonese_llm_v2_chat_gsm8k_GRPO_Qwen3-8B
8B • Updated • 11 • 1
jed351/cantonese_llm_v2_chat_gsm8k_GRPO_Qwen3-30B-A3B-Think-2507
31B • Updated • 8 • 1
jed351/qwen3-30B-A3B-SFT-DPO
Text Generation • 31B • Updated • 14
jed351/qwen3-8B-SFT-DPO
Text Generation • 8B • Updated • 7
jed351/Gemma3-4B-ChatVector_SFT-from-IT_and_IT
4B • Updated • 6
jed351/gemma3-maxtext-conversion-test
4B • Updated • 6
jed351/Gemma3-4B-SFT-from-IT
Image-Text-to-Text • 4B • Updated • 5
jed351/Gemma3-4B-cpt-v2
4B • Updated • 5
datasets 13
jed351/Traditional-Chinese-Common-Crawl-NOT-Cleaned
Viewer • Updated • 1.02B • 1.27k
jed351/Nemotron-3-Nano-RL-Training-Blend-STEM-Yue-Translated
Preview • Updated • 74
jed351/gsm8k_cantonese
Viewer • Updated • 26.4k • 40
jed351/Traditional-Chinese-Common-Crawl-by-year
Viewer • Updated • 15.5M • 353
jed351/Cantonese_Common_Crawl_Filtered
Viewer • Updated • 5.65M • 4.84k • 4
jed351/Traditional-Chinese-Common-Crawl-Filtered
Viewer • Updated • 278M • 2.56k • 26
jed351/Cantonese-Web-Data
Viewer • Updated • 732k • 84 • 5
jed351/fineweb-ja-keyword-hk
Viewer • Updated • 2.08M • 349
jed351/finepdfs-traditional-chinese
Viewer • Updated • 1.31M • 294
jed351/Chinese-Common-Crawl-Filtered
Viewer • Updated • 41.6M • 347 • 18