arxiv:2504.00824
Xueguang Ma PRO
MrLight
AI & ML interests
None yet
Recent Activity
updated
a model
about 8 hours ago
SVRL/verl-scalable-0426-rerun-new-kl_batch768_c0.3_7B_scale-nos-fixmc-fil0-fil8
updated
a model
about 8 hours ago
SVRL/verl-scalable-0426-rerun-new-kl_batch768_c0.3_7B_scale-nos-fixmc-fil0-fil8
updated
a model
about 8 hours ago
SVRL/verl-scalable-0426-rerun-new-kl_batch768_c0.3_7B_scale-nos-fixmc-fil0-fil8
Organizations
Papers
17
models
9
MrLight/qwen3-4B-usekl-newdata_webinstruct-verified_Qwen3-4B
Updated
MrLight/repro-7B-preview-usekl-newdata_webinstruct-verified_Qwen2.5-7B-step300
Updated
MrLight/repro-7B-preview-usekl-newdata_webinstruct-verified_Qwen2.5-7B
Updated
MrLight/repro-7B-preview_scale-reasoning-data-v2-nos-fixmc-fil0-fil8_Qwen2.5-7B
Updated
MrLight/scale-reasoning-data-v2-nos-fixmc-fil0-fil8-Qwen2.5-7B
Updated
•
30
MrLight/temp-3b
Text Generation
•
Updated
MrLight/dse-qwen2-2b-mrl-v1
Visual Document Retrieval
•
Updated
•
5.1k
•
58
MrLight/qwen2-7b-lora-wikifbpub-2e
Updated
•
3
MrLight/dse-phi35-vidore-ft
Updated
•
7
•
10
datasets
12
MrLight/bbeh-eval
Viewer
•
Updated
•
4.52k
•
47
MrLight/v2-qrevised-mathonly-nos-balance-short-filter0-filter8
Viewer
•
Updated
•
56.3k
•
69
MrLight/v2-qrevised-mathonly-nos-filter0-filter8
Viewer
•
Updated
•
103k
•
64
MrLight/qrevised-mathonly-nos-balance-short-filter0-filter8
Viewer
•
Updated
•
56.3k
•
45
MrLight/verl-reasoning-data-v2-qrevised-mathonly-nos-filter0-filter8
Viewer
•
Updated
•
103k
•
53
MrLight/paper-visa
Viewer
•
Updated
•
103k
•
193
MrLight/fineweb-visa
Viewer
•
Updated
•
59.9k
•
135
MrLight/wiki-visa
Viewer
•
Updated
•
90k
•
926
MrLight/wikipedia-20240520
Viewer
•
Updated
•
6.36M
•
37
MrLight/msmarco-doc
Viewer
•
Updated
•
137k
•
28