ItColBERT：意大利语专用后期交互检索器
原标题：ItColBERT: An Italian-Specialised Late-Interaction Retriever
AI 摘要
研究团队发布了 ItColBERT，一个 135M 参数的意大利语专用后期交互检索器，基于 PyLate 并遵循 ColBERT-Zero 训练配方，在单张 RTX 3090 上约 14.5 GPU 小时完成训练。在四个意大利语检索基准上，它优于除 mLateOn 外的所有通用后期交互基线，且参数更少。主要发现是方法学上的：在 MLDR-it 上，对未改动检查点应用推理时分块带来的 nDCG@10 提升大于额外两轮训练；自挖掘硬负例和原生 1024 token 训练均未通过预注册决策门槛。作者开源了权重、训练评估代码及完整实验记录。
正文节选
ItColBERT: An Italian-Specialised Late-Interaction Retriever Abstract Neural information retrieval for Italian is served almost entirely by multilingual models. Several multi-vector (late-interaction) retrievers include Italian among dozens of languages, and several strong Italian dense embedders exist, but as of August 2026 no late-interaction retriever specialised on Italian had been released. We present ItColBERT, a 135M-parameter Italian ColBERT trained with PyLate following the ColBERT-Zero