AI 中文总结
ItColBERT是首个意大利语专用后期交互检索器，以135M参数在四个基准上优于多数通用基线，并发现推理时分块在域外基准上带来显著提升，而额外训练和负样本策略未通过预注册门控。
AI 中文摘要
意大利语的神经信息检索几乎完全由多语言模型提供服务。多个多向量（后期交互）检索器将意大利语纳入其数十种语言之中，并且存在多个强大的意大利语稠密嵌入器，但截至2026年8月，尚未发布专门针对意大利语的后期交互检索器。我们提出了ItColBERT，这是一个135M参数的意大利语ColBERT模型，使用PyLate按照ColBERT-Zero配方进行训练：从已经能够检索的检查点初始化，然后进行监督对比训练，随后进行单教师蒸馏，总计在单个RTX 3090上约14.5 GPU小时。在四个意大利语检索基准上，它优于我们测试的所有通用后期交互基线，除了一个（mLateOn），并且参数数量比除一个规模相当的基线外的所有基线少2-4.4倍。我们的主要实证发现是方法论上的，且部分为负面。在唯一干净的域外基准（MLDR-it）上，对未改变的检查点应用推理时分块配方，nDCG@10提高了+0.0602（p = 0.0225），这一效果大于两轮进一步训练所产生的任何效果。自挖掘的困难负样本和原生1024 token训练均根据预注册的决策门进行评估，且均未通过。我们报告了所有与配对引导检验的比较，并针对经验测量的噪声下限0.0030 nDCG@10，同时我们发布了权重、训练和评估代码以及完整的实验记录，包括被拒绝的轮次。
英文摘要
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 recipe: initialise from a checkpoint that already retrieves, then apply supervised contrastive training followed by single-teacher distillation, for a total of roughly 14.5 GPU-hours on one RTX 3090. Across four Italian retrieval benchmarks it outperforms every general-purpose late-interaction baseline we tested except one (mLateOn), at 2-4.4x fewer parameters than every baseline but one of comparable size. Our principal empirical finding is methodological and partly negative. On the only cleanly out-of-domain benchmark (MLDR-it), an inference-time chunking recipe applied to an unchanged checkpoint yields +0.0602 nDCG@10 (p = 0.0225), a larger effect than anything two further rounds of training produced. Self-mined hard negatives and native 1024-token training were both evaluated against pre-registered decision gates and both failed. We report every comparison with paired bootstrap tests against an empirically measured noise floor of 0.0030 nDCG@10, and we release the weights, the training and evaluation code, and the complete experimental record including the rejected rounds.
