Abstract generation on LongLaMP benchmark leaderboard · SOTA2 Research

Abstract generation on LongLaMP

42.5R1RankGPT (GPT-5-nano)

Evaluation Results

Method	Links
RankGPT (GPT-5-nano)2026.01		42.5	24.5	32.1
Contriever2026.01		42.3	24.2	31.7
PURPLE2026.01		42.3	24.2	32
BM252026.01		42.2	24.2	31.8
RankGPT (Llama-3-8B-Instruct)2026.01		42.2	24.2	31.9
ICR (Llama-3-8B-Instruct)2026.01		41.9	23.9	31.5
RankGPT (GPT-5-nano)2026.01		39.1	22.4	26.9
PURPLE2026.01		38.8	22.1	26.4
BM252026.01		38.6	22	26.2
RankGPT (Llama-3-8B-Instruct)2026.01		38.6	22	26.2
Contriever2026.01		38.5	21.9	26.1
ICR (Llama-3-8B-Instruct)2026.01		38.5	22	26.2
IC-RALM2026.01		38.3	20.4	29.1
REPLUG2026.01		38.2	20.7	28.4
IC-RALM2026.01		37.2	21.1	25.1
REPLUG2026.01		36	21.2	23.8
RankGPT (GPT-5-nano)2026.01		33.6	16.5	21.4
BM252026.01		30.9	15.3	20.8
ICR (Llama-3-8B-Instruct)2026.01		30.7	15.2	20.3
RankGPT (Llama-3-8B-Instruct)2026.01		30.5	15.1	20.4
PURPLE2026.01		30.2	14.9	20
Contriever2026.01		30.1	14.9	19.7
IC-RALM2026.01		29.8	13.6	18.3
PURPLE2026.01		27.9	14.4	17.1
RankGPT (GPT-5-nano)2026.01		27.1	13.7	16.6
BM252026.01		25.8	13.2	16.5
ICR (Llama-3-8B-Instruct)2026.01		25.8	13.2	16.4
Contriever2026.01		25.2	13.1	15.9
RankGPT (Llama-3-8B-Instruct)2026.01		25.1	12.9	15.8
IC-RALM2026.01		25	12.7	15.8
REPLUG2026.01		20.8	11.2	12.5
REPLUG2026.01		15.5	8.8	8.8