List your company

Instruction Following on BBH

67.1AccuracyBaseline Step 0
7.92423.28738.6554.013Feb 5, 2026

Evaluation results

MethodLinks
Baseline Step 0Model=L3.1-8B-Magpie, Chat Template=On, k=12026.0267.11—
StaticModel=L3.1-8B-Magpie, Chat Template=On, k=12026.0257.61—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.9952026.0257.32.4—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.992026.0256.82.6—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.982026.0256.43—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.972026.0256.23.2—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.962026.02563.3—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.952026.0255.63.4—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.92026.0253.83.9—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.872026.0251.94.1—
StaticModel=L3.1-8B-Magpie, Chat Template=On, k=22026.0251.42—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.852026.0250.44.2—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.82026.0245.54.4—
StaticModel=L3.1-8B-Magpie, Chat Template=On, k=32026.0245.13—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.752026.0239.44.6—
Baseline Step 0Model=Qwen3-4B-Inst-2507, Chat Template=Off, k=12026.0234.41—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.72026.0233.64.8—
StaticModel=L3.1-8B-Magpie, Chat Template=On, k=42026.02284—
StaticModel=Qwen3-4B-Inst-2507, Chat Template=Off, k=12026.02281—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.652026.0227.65.1—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.9952026.0227.42.1—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.992026.0227.22.3—
StaticModel=Qwen3-4B-Inst-2507, Chat Template=Off, k=22026.0226.92—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.982026.0226.82.5—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.972026.0226.42.6—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.962026.0226.22.6—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.952026.0226.22.7—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.92026.0226.13—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.872026.0225.83.1—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.852026.0225.23.1—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.82026.0224.63.3—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.752026.0223.53.4—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.72026.0222.53.6—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.652026.0221.83.7—
ConfAdaptModel=L3.1-8B-Magpie, Chat Template=On, tau (τ)=0.62026.0221.25.3—
StaticModel=Qwen3-4B-Inst-2507, Chat Template=Off, k=32026.02213—
ConfAdaptModel=Qwen3-4B-Inst-2507, Chat Template=Off, tau (τ)=0.62026.0219.74—
StaticModel=L3.1-8B-Magpie, Chat Template=On, k=52026.0216.65—
StaticModel=Qwen3-4B-Inst-2507, Chat Template=Off, k=42026.0214.94—
StaticModel=Qwen3-4B-Inst-2507, Chat Template=Off, k=52026.0210.25—
Bi-LoRABackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——43.45
Bi-LoRACost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.45
DeLoRABackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——42.72
DoRABackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——42.98
Flat-LoRACost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.99
Full FTCost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——44.78
HiRABackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——42.89
LoRABackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——42.82
LoRACost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——42.82
LoRA-ESAMCost=×1.x, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.4
LoRA-LookSAMCost=×1.x, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.28
LoRA-nBARCost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.39
LoRA-oBARCost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.44
LoRA-SAMCost=×2, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.49
PiSSABackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——43.25
ROPCost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.47
RWP_fullCost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——43.35
RWP_LoRACost=×1, Backbone=Llama 3.1-8B, Evaluation Protocol=Fine-tuning2025.08——44.1
S^2-SAMBackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——42.44
WSAMBackbone=Llama 3.1-8B, Evaluation Protocol=fine-tuning2025.08——43.16