LoRA adapters trained for 5 progressively shorter chain-of-thought styles on GSM8K, plus the eval artifacts behind the Pareto curve.
Frolov Anatolii
ssurface
·
AI & ML interests
None yet
Recent Activity
updated a dataset about 8 hours ago
s-nlp/toolace-rus-gemma4-31b published a dataset about 8 hours ago
s-nlp/toolace-rus-gemma4-31b updated a model 15 days ago
s-nlp/tool-calling-hallucination-modernbert-base-glaive-100pctOrganizations
models 12
ssurface/tool-calling-hallucination-modernbert-base-unified-final
Token Classification • 0.1B • Updated • 60
ssurface/tool-calling-hallucination-modernbert-base-glaive-100pct
Token Classification • 0.1B • Updated • 45
ssurface/tool-calling-hallucination-modernbert-large-glaive-100pct
Token Classification • 0.4B • Updated • 52
ssurface/qwen3-4b-cot-compress-l5
Text Generation • Updated • 29
ssurface/qwen3-4b-cot-compress-l4
Text Generation • Updated • 28
ssurface/qwen3-4b-cot-compress-l3
Text Generation • Updated • 28
ssurface/qwen3-4b-cot-compress-l2
Text Generation • Updated • 22
ssurface/qwen3-4b-cot-compress-l1
Text Generation • Updated • 26
ssurface/paradetox-gemma-4b-bnb
Updated
ssurface/qwen2.5-1.5B-DimABSA2026-LORA
Text Generation • Updated • 1