List your company

Question Answering on MuSiQue (EM, ASR)

38Exact Match (EM)Clean (No Attack)
-0.489.5119.529.49Jul 1, 2026

Evaluation results

MethodLinks
Clean (No Attack)Framework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.0738—
PARADOXFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07366
PARADOXFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.073511
Clean (No Attack)Framework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.0732—
Fake CompletionFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07301
NaiveFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.072817
PARADOXFramework=WebThinker, LLM Backbone=QwQ-32B2026.07285
NaiveFramework=WebThinker, LLM Backbone=QwQ-32B2026.07266
Clean (No Attack)Framework=WebThinker, LLM Backbone=QwQ-32B2026.0723—
IgnoreFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.072217
IgnoreFramework=WebThinker, LLM Backbone=QwQ-32B2026.07224
NaiveFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07211
PoisonedRAGFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07204
CombinedFramework=WebThinker, LLM Backbone=QwQ-32B2026.07194
TopicFramework=WebThinker, LLM Backbone=QwQ-32B2026.07191
IgnoreFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07191
CombinedFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07190
NaiveFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.071831
Fake CompletionFramework=WebThinker, LLM Backbone=QwQ-32B2026.071818
PARADOXFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07187
Clean (No Attack)Framework=WebThinker, LLM Backbone=DeepSeek-32B2026.0717—
TopicFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07172
IgnoreFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.071621
CombinedFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07162
Fake CompletionFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07161
TopicFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07141
PoisonedRAGFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.071310
Fake CompletionFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.071261
CombinedFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07946
TopicFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07824
KidnapRAGFramework=ReAct, LLM Backbone=Qwen2.5-32B-Inst2026.07866
PoisonedRAGFramework=WebThinker, LLM Backbone=QwQ-32B2026.0785
PoisonedRAGFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.0787
KidnapRAGFramework=WebThinker, LLM Backbone=DeepSeek-32B2026.07853
KidnapRAGFramework=WebThinker, LLM Backbone=QwQ-32B2026.07757
KidnapRAGFramework=ReAct, LLM Backbone=Llama-3.3-70B-Inst2026.07178