Error Attribution on GSM8k Right Dense
64Removal RateLeft Filtering
Evaluation Results
| Method | Links | |
|---|---|---|
| Left FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 64 | |
| VanillaScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.05 | 63 | |
| Left FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 60 | |
| Left FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 60 | |
| Left FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.05 | 60 | |
| Left FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 53 | |
| Left FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.05 | 52 | |
| Two-WayScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 42 | |
| Two-WayScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 40 | |
| Two-WayScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.05 | 40 | |
| VanillaScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.05 | 34 | |
| CRSVPScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.05 | 29 | |
| CRSVPScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 27 | |
| VanillaScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 25 | |
| CRSVPScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.05 | 24 | |
| VanillaScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 22 | |
| CRSVPScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 22 | |
| CRSVPScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 21 | |
| Two-WayScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 21 | |
| Right FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 21 | |
| Right FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 20 | |
| Right FilteringScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 20 | |
| Two-WayScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 20 | |
| Right FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=DyLAN2026.05 | 20 | |
| Two-WayScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.05 | 19 | |
| VanillaScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 17 | |
| VanillaScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=DyLAN2026.05 | 16 | |
| CRSVPScoring Function Type=Naive LLM, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 10 | |
| Right FilteringScoring Function Type=Role-prompted, Backbone=gpt-4o-mini, Agentic Framework=MACNET2026.05 | 10 | |
| Right FilteringScoring Function Type=Fine-tuned, Backbone=Qwen3-1.7B, Agentic Framework=MACNET2026.05 | 10 |