Movie Recommendation on LaMP Movie
57AccuracyPURPLE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PURPLELLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 57 | 49.9 | |
| RankGPT (Llama-3-8B-Instruct)LLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 56.9 | 48.8 | |
| ICR (Llama-3-8B-Instruct)LLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 56.8 | 49.6 | |
| ContrieverLLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 56.6 | 49.5 | |
| RankGPT (GPT-5-nano)LLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 55.3 | 48.2 | |
| BM25LLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 54.5 | 47.2 | |
| REPLUGLLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 51.7 | 44 | |
| PURPLELLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 49.6 | 41.6 | |
| IC-RALMLLM Backbone=Llama-3-70B-Instruct (70.6B)2026.01 | 49 | 41.8 | |
| ICR (Llama-3-8B-Instruct)LLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 48 | 39.3 | |
| ContrieverLLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 47.2 | 39.1 | |
| RankGPT (Llama-3-8B-Instruct)LLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 47.2 | 38.4 | |
| BM25LLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 45.2 | 36.5 | |
| RankGPT (GPT-5-nano)LLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 45.1 | 36.2 | |
| REPLUGLLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 42 | 31.8 | |
| PURPLELLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 38.6 | 34.2 | |
| IC-RALMLLM Backbone=Llama-3-8B-Instruct (8.03B)2026.01 | 38 | 29.9 | |
| REPLUGLLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 37.4 | 32.1 | |
| ContrieverLLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 36 | 31.1 | |
| RankGPT (GPT-5-nano)LLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 35.5 | 31.4 | |
| BM25LLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 34.5 | 29.6 | |
| RankGPT (Llama-3-8B-Instruct)LLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 34.2 | 28.9 | |
| ICR (Llama-3-8B-Instruct)LLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 34.1 | 29.8 | |
| IC-RALMLLM Backbone=Phi-4-Mini-Instruct (3.84B)2026.01 | 33.6 | 28.8 |