List your company

Error Attribution on GSM8k Mid Dense

75Removal RateVanilla
12.628.84561.2May 7, 2026

Evaluation results

MethodLinks
VanillaScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0575
VanillaScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0566
Two-WayScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0563
Two-WayScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0562
Two-WayScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0560
Two-WayScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0559
Two-WayScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0559
Two-WayScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0559
Left FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0553
Left FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0547
Right FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0545
Right FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0544
Right FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0543
Right FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0543
CRSVPScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0538
Left FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0536
CRSVPScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0536
Left FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0535
Right FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0533
Right FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0533
Left FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0531
Left FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0531
CRSVPScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0529
VanillaScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0527
VanillaScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0527
CRSVPScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0527
CRSVPScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0526
CRSVPScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0521
VanillaScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0518
VanillaScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0515