E-commerce Navigation and Search on WebShop (semantic shift Hidden drift)
100ScoreNo Memory (Plain)
Evaluation Results
| Method | Links | |
|---|---|---|
| No Memory (Plain)Backbone=Qwen3-30B, Agent Architecture=No Memory, GLOVE Augmentation=false2026.01 | 100 | |
| VanillaBackbone=Qwen3-30B, Agent Architecture=Vanilla, GLOVE Augmentation=false2026.01 | 100 | |
| Vanilla + GLOVEBackbone=Qwen3-30B, Agent Architecture=Vanilla, GLOVE Augmentation=true2026.01 | 100 | |
| MemoryBankBackbone=Qwen3-30B, Agent Architecture=MemoryBank, GLOVE Augmentation=false2026.01 | 100 | |
| MemoryBank + GLOVEBackbone=Qwen3-30B, Agent Architecture=MemoryBank, GLOVE Augmentation=true2026.01 | 100 | |
| VoyagerBackbone=Qwen3-30B, Agent Architecture=Voyager, GLOVE Augmentation=false2026.01 | 100 | |
| Voyager + GLOVEBackbone=Qwen3-30B, Agent Architecture=Voyager, GLOVE Augmentation=true2026.01 | 100 | |
| Generative AgentBackbone=Qwen3-30B, Agent Architecture=Generative Agent, GLOVE Augmentation=false2026.01 | 100 | |
| Generative Agent + GLOVEBackbone=Qwen3-30B, Agent Architecture=Generative Agent, GLOVE Augmentation=true2026.01 | 100 | |
| MemoryBank + GLOVELLM Backbone=GPT-4o, Base Agent=MemoryBank2026.01 | 98.8 | |
| Voyager + GLOVELLM Backbone=GPT-4o, Base Agent=Voyager2026.01 | 98.8 | |
| Vanilla +GLOVELLM=Grok-3, Augmentation=GLOVE2026.01 | 98.8 | |
| MemoryBank +GLOVELLM=Grok-3, Augmentation=GLOVE2026.01 | 98.8 | |
| Voyager +GLOVELLM=Grok-3, Augmentation=GLOVE2026.01 | 98.8 | |
| Vanilla + GLOVEBackbone=DeepSeek-V3.22026.01 | 97.5 | |
| MemoryBank + GLOVEBackbone=DeepSeek-V3.22026.01 | 97.5 | |
| Voyager + GLOVEBackbone=DeepSeek-V3.22026.01 | 97.5 | |
| Generative Agent + GLOVEBackbone=DeepSeek-V3.22026.01 | 97.5 | |
| Generative Agent +GLOVELLM=Grok-3, Augmentation=GLOVE2026.01 | 97.5 | |
| No Memory (Plain)LLM=Grok-32026.01 | 96.2 | |
| Voyager + GLOVEBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 95 | |
| No Memory (Plain)Backbone=DeepSeek-V3.22026.01 | 95 | |
| Vanilla + GLOVELLM Backbone=GPT-4o, Base Agent=Vanilla2026.01 | 93.8 | |
| MemoryBank + GLOVEBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 88.8 | |
| No Memory (Plain)LLM Backbone=GPT-4o2026.01 | 87.5 | |
| MemoryBankBackbone=DeepSeek-V3.22026.01 | 82.5 | |
| Generative AgentLLM=Grok-32026.01 | 82.5 | |
| MemoryBankBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 81.2 | |
| MemoryBankLLM Backbone=GPT-4o2026.01 | 81.2 | |
| MemoryBankLLM=Grok-32026.01 | 81.2 | |
| No Memory (Plain)Backbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 80 | |
| Vanilla + GLOVEBackbone=Llama3.1-8B2026.01 | 76.2 | |
| VanillaBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 75 | |
| Vanilla + GLOVEBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 75 | |
| VoyagerBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 75 | |
| Generative AgentBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 75 | |
| Generative Agent + GLOVEBackbone=Llama3.3-70B, Drift Type=Implicit, Stage=Hidden drift2026.01 | 75 | |
| VanillaBackbone=Qwen2.5-7B2026.01 | 75 | |
| Vanilla + GLOVEBackbone=Qwen2.5-7B2026.01 | 75 | |
| MemoryBank + GLOVEBackbone=Qwen2.5-7B2026.01 | 75 | |
| VoyagerBackbone=Qwen2.5-7B2026.01 | 75 | |
| Voyager + GLOVEBackbone=Qwen2.5-7B2026.01 | 75 | |
| Generative AgentBackbone=Qwen2.5-7B2026.01 | 75 | |
| Generative Agent + GLOVEBackbone=Qwen2.5-7B2026.01 | 75 | |
| VanillaBackbone=DeepSeek-V3.22026.01 | 75 | |
| VoyagerBackbone=DeepSeek-V3.22026.01 | 75 | |
| Generative AgentBackbone=DeepSeek-V3.22026.01 | 75 | |
| VanillaLLM Backbone=GPT-4o2026.01 | 75 | |
| VoyagerLLM Backbone=GPT-4o2026.01 | 75 | |
| Generative AgentLLM Backbone=GPT-4o2026.01 | 75 | |
| Generative Agent + GLOVELLM Backbone=GPT-4o, Base Agent=Generative Agent2026.01 | 75 | |
| VanillaLLM=Grok-32026.01 | 75 | |
| VoyagerLLM=Grok-32026.01 | 75 | |
| MemoryBank + GLOVEBackbone=Llama3.1-8B2026.01 | 67.5 | |
| No Memory (Plain)Backbone=Llama3.1-8B2026.01 | 60 | |
| Generative Agent + GLOVEBackbone=Llama3.1-8B2026.01 | 60 | |
| Generative AgentBackbone=Llama3.1-8B2026.01 | 55 | |
| MemoryBankBackbone=Qwen2.5-7B2026.01 | 53.8 | |
| Voyager + GLOVEBackbone=Llama3.1-8B2026.01 | 42.5 | |
| VanillaBackbone=Llama3.1-8B2026.01 | 36.2 | |
| MemoryBankBackbone=Llama3.1-8B2026.01 | 27.5 | |
| VoyagerBackbone=Llama3.1-8B2026.01 | 26.2 | |
| No Memory (Plain)Backbone=Qwen2.5-7B2026.01 | 15 |