List your company

Question Answering on 2WikiMultihopQA (Guard Rate)

100Guard RateIgnore
-4235077Jul 1, 2026

Evaluation results

MethodLinks
IgnoreFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07100
CombinedFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07100
PoisonedRAGFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07100
IgnoreFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07100
CombinedFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07100
PoisonedRAGFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.0799
Fake CompletionFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.0798.9
Fake CompletionFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.0798.8
NaiveFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.0798.7
NaiveFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.0797
PARADOXFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.0795.9
PARADOXFramework=WebThinker, LLM Backbone=QwQ-32B2026.0793.5
PARADOXFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.0789
TopicFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.0777.2
TopicFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.0775.8
PARADOXFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0768
PoisonedRAGFramework=WebThinker, LLM Backbone=QwQ-32B2026.0745.7
NaiveFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0740.5
TopicFramework=WebThinker, LLM Backbone=QwQ-32B2026.0738
Fake CompletionFramework=WebThinker, LLM Backbone=QwQ-32B2026.0734.9
TopicFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0732.7
PoisonedRAGFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0728.9
NaiveFramework=WebThinker, LLM Backbone=QwQ-32B2026.0728.2
IgnoreFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0727.8
Fake CompletionFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0726.7
CombinedFramework=WebThinker, LLM Backbone=QwQ-32B2026.0724.4
CombinedFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0723.3
IgnoreFramework=WebThinker, LLM Backbone=QwQ-32B2026.0716.9
KidnapRAGFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.078
KidnapRAGFramework=WebThinker, LLM Backbone=QwQ-32B2026.073.1
KidnapRAGFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.070
KidnapRAGFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.070