dorsal/arxiv
View SchemaAn Efficient Long-Context Ranking Architecture With Calibrated LLM Distillation: Application to Person-Job Fit
| Authors | Warren Jouanneau, Emma Jouffroy, Marc Palyart |
|---|---|
| Categories | |
| ArXiv ID | 2601.10321vv2 |
| URL | https://arxiv.org/abs/2601.10321 |
| License | http://creativecommons.org/licenses/by/4.0/ |
Abstract
Finding the most relevant person for a job proposal in real time is challenging, especially when resumes are long, structured, and multilingual. In this paper, we propose a re-ranking model based on a new generation of late cross-attention architecture, that decomposes both resumes and project briefs to efficiently handle long-context inputs with minimal computational overhead. To mitigate historical data biases, we use a generative large language model (LLM) as a teacher, generating fine-grained, semantically grounded supervision. This signal is distilled into our student model via an enriched distillation loss function. The resulting model produces skill-fit scores that enable consistent and interpretable person-job matching. Experiments on relevance, ranking, and calibration metrics demonstrate that our approach outperforms state-of-the-art baselines.
{
"annotation_id": "c5f871d6-b84c-4a29-aba2-df34d4288b0e",
"date_created": "2026-02-17T05:53:24.292000Z",
"date_modified": "2026-02-17T05:53:24.292000Z",
"file_hash": "37d34cf756d3ca2d55002960ed88df4fd51bdf61dc38be60f0658e0b8fb6c490",
"private": false,
"record": {
"abstract": "Finding the most relevant person for a job proposal in real time is challenging, especially when resumes are long, structured, and multilingual. In this paper, we propose a re-ranking model based on a new generation of late cross-attention architecture, that decomposes both resumes and project briefs to efficiently handle long-context inputs with minimal computational overhead. To mitigate historical data biases, we use a generative large language model (LLM) as a teacher, generating fine-grained, semantically grounded supervision. This signal is distilled into our student model via an enriched distillation loss function. The resulting model produces skill-fit scores that enable consistent and interpretable person-job matching. Experiments on relevance, ranking, and calibration metrics demonstrate that our approach outperforms state-of-the-art baselines.",
"arxiv_id": "2601.10321",
"authors": [
"Warren Jouanneau",
"Emma Jouffroy",
"Marc Palyart"
],
"categories": [
"cs.CL",
"cs.IR",
"cs.LG",
"cs.SI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"title": "An Efficient Long-Context Ranking Architecture With Calibrated LLM Distillation: Application to Person-Job Fit",
"url": "https://arxiv.org/abs/2601.10321",
"version": "v2"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "23bd5c77-ece2-49e8-8989-b0d4cfe5696c",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}