Vietnamese passage reranker
Hub tags mark the model as a Vietnamese cross-encoder rerank model, and the model card documents CrossEncoder scoring over word-segmented query-passage pairs.
Open Source Model Profile · itdainb
PhoRanker
PhoRanker is a 135M-parameter RoBERTa cross-encoder from itdainb for Vietnamese passage reranking with 256-token CrossEncoder use.
- Publisher
- itdainb
- Task
- text-classification
- Model type
- roberta
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
PhoRanker is published by itdainb as a RoBERTa text-classification rerank model. The captured configuration identifies RobertaForSequenceClassification and Safetensors metadata reports 134,999,041 parameters. Hub tags describe Vietnamese cross-encoder reranking, and card data records apache-2.0.
Recorded capabilities
VnCoreNLP preprocessing
According to the model card, text is word-segmented with VnCoreNLP before scoring, with sentence-transformers and Transformers examples.
Reported Vi reranking table
The model card reports 0.7422 NDCG@10 and 0.6830 MRR@10 on the Vietnamese MS MARCO dev set, with A100 fp16 runtime of 15 docs per second.
Use cases in the source record
- Vietnamese passage reranking that scores word-segmented candidate passages for a query with the documented CrossEncoder workflow.
- Transformers sequence-classification scoring with sigmoid outputs, following the publisher's fp16-compatible loading example.
Limitations and unknowns
- No independent evaluation results were extracted; performance figures are publisher-reported model-card values.
- No context-window value was extracted beyond the documented 256-token usage setting.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: itdainb/PhoRanker
Captured: Unknown. Processed: 2026-09-07T19:34:47.462848+00:00.
Table of contents Installation Pre-processing Usage with sentence-transformers Usage with transformers Performance Support me Citation Installation Install VnCoreNLP to word segment: pip install py_vncorenlp Install sentence-transformers (recommend) - Usage : pip install sentence-transformers Install transformers (optional) - Usage : pip install transformers Pre-processing import py_vncorenlp py_vncorenlp.download_model(save_dir= '/absolute/path/to/vncorenlp' ) rdrsegmenter = py_vncorenlp.VnCoreNLP(annotators=[ "wseg" ], save_dir= '/absolute/path/to/vncorenlp' ) query = "Trường UIT là gì?" sentences = [ "Trường Đại học Công nghệ Thô…
F001F002F003F004F005F006F007F009F010F013F014F015F016F017F018F019