dense-index
RAVine-dense-index
RAVine-dense-index
This repository contains dense index files for the search tools of the RAVine: Reality-Aligned Evaluation for Agentic Search framework. The corpus is MS MARCO V2.1, encoded using Alibaba-NLP/gte-modernbert-base.
Paper: RAVine: Reality-Aligned Evaluation for Agentic Search
Code: https://github.com/SwordFaith/RAVine
Abstract
Agentic search, as a more autonomous and adaptive paradigm of retrieval augmentation, is driving the evolution of intelligent… See the full description on the dataset page: https://huggingface.co/datasets/sapphirex/RAVine-dense-index.denseIndexlongeval-2026-snapshot-2-index-dense
LongEval 2026 Snapshot 2 Terrier Dense Index
Description
This is a Terrier Dense index for the snapshot-2 of the LongEval 2026 test collection. It indexes the titles and abstracts of the scientific documents using Qwen/Qwen3-Embedding-4B and the default prompt.
# Load the artifact
import pyterrier as pt
import pyterrier_dr
index = pt.Artifact.from_hf('jueri/longeval-2026-snapshot-2-index-dense')
model = pyterrier_dr.SBertBiEncoder('Qwen/Qwen3-Embedding-4B')
retriever =… See the full description on the dataset page: https://huggingface.co/datasets/jueri/longeval-2026-snapshot-2-index-dense.longeval-2026-snapshot-1-index-dense
LongEval 2026 Snapshot 1 Terrier Dense Index
Description
This is a Terrier Dense index for the snapshot-1 of the LongEval 2026 test collection. It indexes the titles and abstracts of the scientific documents using Qwen/Qwen3-Embedding-4B and the default prompt.
# Load the artifact
import pyterrier as pt
import pyterrier_dr
index = pt.Artifact.from_hf('jueri/longeval-2026-snapshot-1-index-dense')
model = pyterrier_dr.SBertBiEncoder('Qwen/Qwen3-Embedding-4B')
retriever =… See the full description on the dataset page: https://huggingface.co/datasets/jueri/longeval-2026-snapshot-1-index-dense.longeval-2026-snapshot-3-index-dense
LongEval 2026 Snapshot 3 Terrier Dense Index
Description
This is a Terrier Dense index for the snapshot-3 of the LongEval 2026 test collection. It indexes the titles and abstracts of the scientific documents using Qwen/Qwen3-Embedding-4B and the default prompt.
# Load the artifact
import pyterrier as pt
import pyterrier_dr
index = pt.Artifact.from_hf('jueri/longeval-2026-snapshot-3-index-dense')
model = pyterrier_dr.SBertBiEncoder('Qwen/Qwen3-Embedding-4B')
retriever =… See the full description on the dataset page: https://huggingface.co/datasets/jueri/longeval-2026-snapshot-3-index-dense.