Modelpublic

Jnx03/kanitakorn-v57-clean-exam-skills-lr1e7-c1-final

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face

0likes3downloads

Model Card

Kanitakorn 2026-06-13 campaign artifact.

—Base: deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
—Version: v57-clean-exam-skills-lr1e7-c1-final
—Method: sft-lora
—Final-claim policy: one model, no BoN, no self-consistency, no ensemble, no model routing.
—Date: 2026-06-15

Benchmark	Score	Target	Status
aime24_th	pending	>15.0	pending
aime24	pending	>25.0	pending
math500_th	pending	>56.0	pending
math500	pending	>82.0	pending
livecodebench_th	pending	>35.0	pending
livecodebench	pending	>60.0	pending
openthaieval	pending	>80.0	pending
hotpotqathen	pending	>46.0	pending
instructionfollowingth_en	pending	>57.0	pending
mtbenchth_en	pending	>85.0	pending
thaiexam	pending	>70.0	pending
ifeval_th	pending	>82.0	pending

Automatic checkpoint publish from runs/20260615deepseekv57cleanexamskillsfroma0p90lr1e7_c1/final. Single-model artifact; no BoN/self-consistency.