List your company

Personalized Question Answering on LaMP-QA (test)

49.18Art ScoreIAP(Ours)
24.06430.584537.10543.6255Aug 14, 2025Sep 28, 2025Nov 12, 2025Dec 27, 2025Feb 10, 2026Mar 27, 2026May 12, 2026

Evaluation results

MethodLinks
IAP(Ours)Base Model=Mistral-Small 24B2026.0549.1866.4966.28—60.65
IAP(Ours)Base Model=Mistral-Small 24B-it2026.0548.1165.0565.36—59.5
IAP(Ours)Base Model=Gemma3-12B-it2026.0547.6959.3466.65—57.89
IAP(Ours)Base Model=Gemma3-12B2026.0546.1262.1864.3—57.53
SFTBase Model=Mistral-Small 24B2026.0543.5962.4764.3—56.78
SFTBase Model=Gemma3-12B-it2026.0542.8958.0161.18—54.02
Explicit Intent Pers.Base Model=Gemma3-12B-it2026.0542.759.8260.88—54.46
PR2Base LLM=Qwen2.5-7B2026.0242.6555.1959.8352.56—
SFTBase Model=Gemma3-12B2026.0542.361.4461.47—55.07
Implicit Intent Pers.Base Model=Gemma3-12B-it2026.0542.1158.7260.44—53.75
Explicit Intent Pers.Base Model=Mistral-Small 24B2026.0541.8962.1262.55—55.52
SFTBase Model=Mistral-Small 24B-it2026.0541.6158.7861.57—53.98
Implicit Intent Pers.Base Model=Mistral-Small 24B2026.0541.5860.3361.84—54.58
Explicit Intent Pers.Base Model=Gemma3-12B2026.0541.5561.2961.38—54.74
Non-pers.Base Model=Gemma3-12B-it2026.0541.359.5262.92—54.58
Implicit Intent Pers.Base Model=Gemma3-12B2026.0540.7860.3461.2—54.1
Explicit Intent Pers.Base Model=Mistral-Small 24B-it2026.0540.4757.8860.72—53.02
Non-pers.Base Model=Gemma3-12B2026.0540.3360.0560.48—53.62
Implicit Intent Pers.Base Model=Mistral-Small 24B-it2026.0540.2857.3860.49—52.71
Non-pers.Base Model=Mistral-Small 24B2026.0540.2758.1259.26—52.55
Non-pers.Base Model=Mistral-Small 24B-it2026.0539.6656.160.03—51.93
PR2Base LLM=Gemma3-4B2026.0238.5552.7157.3249.52—
PrLMBase LLM=Qwen2.5-7B2026.0238.4948.0854.2246.93—
PR2Base LLM=Qwen2.5-3B2026.0237.5149.2457.6148.12—
Search R1-GRPOBase LLM=Qwen2.5-7B2026.0236.6849.3954.1646.74—
IAP(Ours)Base Model=Qwen2.5-7B2026.0536.2948.8951.78—45.65
HYDRA-AdapterBase LLM=Qwen2.5-7B2026.0236.1247.8653.1945.72—
Offline RL RAG-PersonalizationRuntime (second / query)=1.672025.0835.7946.2150.744.23—
IAP(Ours)Base Model=Qwen2.5-7B-it2026.0535.3347.1852.4—44.97
Search R1-GRPOBase LLM=Gemma3-4B2026.0235.2747.3152.6645.08—
PlanPersRuntime (second / query)=3.122025.0835.1848.1852.445.25—
PPlugBase LLM=Qwen2.5-7B2026.0234.8847.7353.0145.2—
PPlugBase LLM=Gemma3-4B2026.0234.5943.6251.4143.2—
VACRuntime (second / query)=1.632025.0834.5452.8853.3146.91—
Search R1-GRPOBase LLM=Qwen2.5-3B2026.0234.1146.9151.444.14—
RAG-PersonalizationRuntime (second / query)=1.712025.0833.9744.8149.6742.81—
PPlugBase LLM=Qwen2.5-3B2026.0233.7445.2250.7743.24—
HYDRA-AdapterBase LLM=Gemma3-4B2026.0232.7146.2552.7343.89—
PrLMBase LLM=Gemma3-4B2026.0232.6648.1753.0144.61—
PrLMBase LLM=Qwen2.5-3B2026.0232.2847.6252.7644.22—
SFTBase Model=Qwen2.5-7B2026.0532.2145.2648.37—41.94
Explicit Intent Pers.Base Model=Qwen2.5-7B2026.0531.6946.5848.03—42.1
SFTBase Model=Qwen2.5-7B-it2026.0531.5843.1148.01—40.9
Implicit Intent Pers.Base Model=Qwen2.5-7B2026.0531.5846.3947.77—41.91
No-PersonalizationRuntime (second / query)=0.782025.0831.2945.8247.6941.6—
Non-pers.Base Model=Qwen2.5-7B2026.0531.2945.8247.69—41.6
RAG-PersonalizationBase LLM=Qwen2.5-7B2026.0230.8641.0745.939.28—
Explicit Intent Pers.Base Model=Qwen2.5-7B-it2026.0530.3242.6647.66—40.21
Implicit Intent Pers.Base Model=Qwen2.5-7B-it2026.0530.0442.0647.29—39.79
No personalizationBase LLM=Qwen2.5-7B2026.0229.8140.1443.5237.82—
HYDRA-AdapterBase LLM=Qwen2.5-3B2026.0229.4245.0452.1142.19—
Non-pers.Base Model=Qwen2.5-7B-it2026.0529.2940.8344.69—38.27
No personalizationBase LLM=Gemma3-4B2026.0229.1738.7144.6237.5—
RAG-PersonalizationBase LLM=Gemma3-4B2026.0228.8739.6542.3436.95—
RAG-PersonalizationBase LLM=Qwen2.5-3B2026.0227.5337.2842.8235.88—
RAG-Personalization (Random P)Runtime (second / query)=1.652025.0825.4738.2940.3734.71—
No personalizationBase LLM=Qwen2.5-3B2026.0225.0335.1140.7333.62—