Preference Optimization for Implicit Model Fusion
Fanqi Wan
Wanfq
AI & ML interests
Large Language Models, Model Fusion, Self-Improving, Instruction-Tuning, Hallucination Mitigation, Dialogue Systems
Recent Activity
liked
a model
5 days ago
FuseAI/FuseChat-Gemma-2-9B-Instruct
liked
a model
5 days ago
FuseAI/FuseChat-Qwen-2.5-7B-Instruct
liked
a model
5 days ago
FuseAI/FuseChat-Llama-3.2-1B-Instruct
Organizations
Collections
7
models
66
Wanfq/FuseLLM-7B
Text Generation
•
Updated
•
551
•
21
Wanfq/fusechat_v1_multi_teacher_woref_fnan_ckpt
Text Generation
•
Updated
•
9
Wanfq/fusechat_v1_nous_hermes_mixtral_teacher_fnan_ckpt
Text Generation
•
Updated
•
9
Wanfq/fusechat_v1_nous_hermes_solar_teacher_fnan_ckpt
Text Generation
•
Updated
•
9
Wanfq/fusechat_v1_multi_teacher_fnan_ckpt
Text Generation
•
Updated
•
9
Wanfq/KCA_Llama_2_13B_Refusal_Tuning
Text Generation
•
Updated
•
12
Wanfq/KCA_Llama_2_13B_Discarding_Tuning
Text Generation
•
Updated
•
15
Wanfq/KCA_Llama_2_13B_Open-Book_Tuning
Text Generation
•
Updated
•
10
Wanfq/KCA_Pythia_6.9B_Refusal_Tuning
Text Generation
•
Updated
•
17
Wanfq/KCA_Pythia_6.9B_Discarding_Tuning
Text Generation
•
Updated
•
14
datasets
32
Wanfq/3_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representation
Viewer
•
Updated
•
21.8k
•
57
Wanfq/3_4_fusechat_v1_openchat-3.5_nh2-mixtral-8x7b-dpo_nh2-solar-10.7b_representation
Preview
•
Updated
•
42
Wanfq/2_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representation
Viewer
•
Updated
•
21.8k
•
34
Wanfq/2_4_fusechat_v1_openchat-3.5_nh2-mixtral-8x7b-dpo_nh2-solar-10.7b_representation
Preview
•
Updated
•
40
Wanfq/1_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representation
Viewer
•
Updated
•
21.7k
•
26
Wanfq/1_4_fusechat_v1_openchat-3.5_nh2-mixtral-8x7b-dpo_nh2-solar-10.7b_representation
Preview
•
Updated
•
23
Wanfq/0_4_fusechat_v1_openchat-3.5_mixtral-8x7b-instruct-v0.1_solar-10.7b-instruct-v1.0_representation
Preview
•
Updated
•
33
Wanfq/0_4_fusechat_v1_openchat-3.5_nh2-mixtral-8x7b-dpo_nh2-solar-10.7b_representation
Preview
•
Updated
•
32
Wanfq/KCA_data
Preview
•
Updated
•
270
•
3
Wanfq/wizardcoder
Viewer
•
Updated
•
57.4k
•
14
•
2