๐—ง๐—ฒ๐—ฎ๐—ฐ๐—ต๐—ถ๐—ป๐—ด ๐—” ๐—ฅ๐—ฒ๐—ฟ๐—ฎ๐—ป๐—ธ๐—ฒ๐—ฟ ๐—ง๐—ต๐—ฒ ๐—Ÿ๐—ฎ๐—ป๐—ด๐˜‚๐—ฎ๐—ด๐—ฒ ๐—ข๐—ณ ๐—ฆ๐—ฒ๐—ฐ๐˜‚๐—ฟ๐—ถ๐˜๐˜† ๐—ง๐—ถ๐—ฐ๐—ธ๐—ฒ๐˜๐˜€

We improved our RAG pipeline. We fine-tuned a reranker for security tickets. The result was a 41 percent increase in MRR@10. The score rose from 0.598 to 0.846.

We kept the model architecture. We kept the embedding model. We trained the reranker on our own data.

We found training data in 142,000 closed tickets. Analysts often write "Refer to ticket #123". These notes are free relevance labels. They show related tickets.

Our method:

Hard negatives matter most. These items look right to the embedder but are wrong. Teaching the model these gaps increases accuracy.

Use this approach if:

Build your evaluation tool on day one. Find labels in your existing text.

Source: https://dev.to/vinayiitkgp/teaching-a-reranker-the-language-of-security-tickets-41-mrr10-4mgk Optional learning community: https://t.me/GyaanSetuAi