arxiv:2504.10766
Yanhong Li
yanhong-li
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 17 hours ago
Text or Pixels? It Takes Half: On the Token Efficiency of Visual Text
Inputs in Multimodal LLMs
published
a model
5 days ago
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL35_selection
published
a model
5 days ago
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL1_selection
Organizations
None yet
models
14
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL35_selection
Updated
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL1_selection
Updated
yanhong-li/qwen2_3b_instruct_gla_v1_hybrid_0_5_ppl_selection
Updated
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_mse_selection
Updated
yanhong-li/qwen2_3b_instruct_gla_v1_hybrid_0_33_uniform
Updated
yanhong-li/qwen2_3b_instruct_gdn_v4_hybrid_30attn_trained_s2_gdnv4_num_layer_35_selection
Updated
yanhong-li/llama3_3b_gdn_v4_hybrid_22attn_trained_s2_gdnv4_num_layer_35_selection
Updated
yanhong-li/llama3_3b_gdn_v4_hybrid_0_5_kv_selection
Updated
yanhong-li/llama3_3b_gdn_v4_hybrid_0_125_S1_LL1_35_selection
Updated
yanhong-li/qwen2_3b_instruct_gla_v1_hybrid_0_25_ar_mutihop_selection
Updated
datasets
0
None public yet