List your company

Knowledge-intensive Claim Extraction and Information Retrieval on 75-case (test)

3.555Overall ScoreDisasterLex
-0.070440.870781.8122.75322May 28, 2026

Evaluation results

MethodLinks
DisasterLexBase Model=Qwen 3.6 Flash2026.053.5553.5543.6792.9624.564
DisasterLexBase Model=Gemini 3.1 Flash Lite Preview2026.053.5423.6763.6072.9524.449
DisasterLexBase Model=DeepSeek V3.22026.053.0333.0443.2352.2924.256
DisasterLexBase Model=Qwen3 8B2026.052.5662.5982.4921.923.923
DisasterLexBase Model=Qwen3 32B2026.052.5652.6572.4661.9743.769
DisasterLexBase Model=Llama 3.1 8B2026.051.7531.8641.4051.3943.03
LightRAGBase Model=Qwen3 8B2026.051.751.0442.4840.6793.744
DisasterLexBase Model=Llama 3.3 70B2026.051.651.6621.1511.3462.974
LightRAGBase Model=Qwen 3.6 Flash2026.051.4070.9411.8010.3973.462
LightRAGBase Model=Qwen3 32B2026.051.3880.6322.2460.5382.821
LightRAGBase Model=Gemini 3.1 Flash Lite Preview2026.051.2860.6761.9650.3882.885
LightRAGBase Model=Llama 3.1 8B2026.051.2680.8241.9450.662.077
HippoRAG 2Base Model=Llama 3.3 70B2026.051.20.3382.0260.3242.872
HippoRAG 2Base Model=Llama 3.1 8B2026.051.1820.5591.7590.263
LightRAGBase Model=DeepSeek V3.22026.051.1230.5881.5340.2213.026
LightRAGBase Model=Llama 3.3 70B2026.051.0760.8091.7060.5381.577
HippoRAG 2Base Model=Qwen3 32B2026.051.040.1761.4390.342.987
HippoRAG 2Base Model=Qwen3 8B2026.050.9070.0591.3890.2692.59
HippoRAG 2Base Model=DeepSeek V3.22026.050.6510.0880.710.092.423
ReFoRCEBase Model=Gemini 3.1 Flash Lite Preview2026.050.620.1570.4740.4551.769
HippoRAG 2Base Model=Qwen 3.6 Flash2026.050.5910.1470.4120.0382.538
ReFoRCEBase Model=Qwen3 32B2026.050.5160.3140.2370.5451.128
ReFoRCEBase Model=DeepSeek V3.22026.050.5030.0690.4560.3491.444
HippoRAG 2Base Model=Gemini 3.1 Flash Lite Preview2026.050.50200.3770.0582.231
ReFoRCEBase Model=Llama 3.3 70B2026.050.4990.1960.3510.2851.538
CHESSBase Model=Qwen 3.6 Flash2026.050.4620.5880.3250.4420.538
CHESSBase Model=DeepSeek V3.22026.050.4340.2350.1840.261.41
CHESSBase Model=Gemini 3.1 Flash Lite Preview2026.050.4170.2750.4560.2790.821
ReFoRCEBase Model=Qwen 3.6 Flash2026.050.35400.4090.081.282
ReFoRCEBase Model=Qwen3 8B, Note=n = 2 seeds2026.050.330.2940.2890.3170.462
CHESSBase Model=Qwen3 32B2026.050.2420.0780.1230.1470.821
CHESSBase Model=Llama 3.1 8B2026.050.2380.0690.360.0320.692
ReFoRCEBase Model=Llama 3.1 8B2026.050.21200.0880.0221.051
CHESSBase Model=Llama 3.3 70B2026.050.1580.0390.0610.1030.564
CHESSBase Model=Qwen3 8B2026.050.06900.070.0960.103