ComprExIT context-compression models and ICAE baselines: 1B/3B/8B, 4x compression, NTP and SFT variants.
Jiangnan Ye
Jiang-nan
AI & ML interests
NLP, Graph Neural Networks, Computer Vision
Recent Activity
upvoted a paper 6 days ago
Aspire: Can Models Self-Evolve from Vague Goals? upvoted a paper about 1 month ago
Recursive Synthesis for Long-Horizon Terminal Tasks upvoted a paper about 1 month ago
Chunky Post-Training: Data Driven Failures of GeneralizationOrganizations
None yet
models 16
Jiang-nan/icae-1b-4x-long8192-sft
3B • Updated • 4
Jiang-nan/icae-1b-4x-long8192-ntp
3B • Updated • 3
Jiang-nan/icae-8b-4x-sft
16B • Updated • 3
Jiang-nan/icae-8b-4x-ntp
8B • Updated • 3
Jiang-nan/icae-3b-4x-sft
7B • Updated • 2
Jiang-nan/icae-3b-4x-ntp
7B • Updated • 4
Jiang-nan/icae-1b-4x-sft
3B • Updated • 4
Jiang-nan/icae-1b-4x-ntp
3B • Updated • 2
Jiang-nan/comprexit-1b-4x-long8192-sft
1B • Updated • 4
Jiang-nan/comprexit-1b-4x-long8192-ntp
1B • Updated • 2