Question Answering on 2WikiMultihopQA (Guard Rate)
100Guard RateIgnore
Evaluation Results
| Method | Links | |
|---|---|---|
| IgnoreFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 100 | |
| CombinedFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 100 | |
| PoisonedRAGFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 100 | |
| IgnoreFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 100 | |
| CombinedFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 100 | |
| PoisonedRAGFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 99 | |
| Fake CompletionFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 98.9 | |
| Fake CompletionFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 98.8 | |
| NaiveFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 98.7 | |
| NaiveFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 97 | |
| PARADOXFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 95.9 | |
| PARADOXFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 93.5 | |
| PARADOXFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 89 | |
| TopicFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 77.2 | |
| TopicFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 75.8 | |
| PARADOXFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 68 | |
| PoisonedRAGFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 45.7 | |
| NaiveFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 40.5 | |
| TopicFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 38 | |
| Fake CompletionFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 34.9 | |
| TopicFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 32.7 | |
| PoisonedRAGFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 28.9 | |
| NaiveFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 28.2 | |
| IgnoreFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 27.8 | |
| Fake CompletionFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 26.7 | |
| CombinedFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 24.4 | |
| CombinedFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 23.3 | |
| IgnoreFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 16.9 | |
| KidnapRAGFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07 | 8 | |
| KidnapRAGFramework=WebThinker, LLM Backbone=QwQ-32B2026.07 | 3.1 | |
| KidnapRAGFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07 | 0 | |
| KidnapRAGFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07 | 0 |