List your company

Error Attribution on GSM8k Left Dense

82Removal RateVanilla
6.0825.7945.565.21May 7, 2026

Evaluation results

MethodLinks
VanillaScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0582
CRSVPScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0580
VanillaScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0578
Right FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0571
CRSVPScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0569
Right FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0565
Right FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0564
Right FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0563
Right FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0563
Right FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0559
CRSVPScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0558
CRSVPScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0554
Left FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0531
VanillaScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0530
CRSVPScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0527
VanillaScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0524
Left FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0524
VanillaScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0524
Two-WayScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0522
VanillaScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0521
CRSVPScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0521
Two-WayScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.0521
Two-WayScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0520
Two-WayScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0518
Two-WayScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.0518
Two-WayScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0517
Left FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0510
Left FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.0510
Left FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.0510
Left FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.059