HybridRAG-BN: A Retrieval-Augmented Framework with Fine-Tuned Verification for Bangla KBQA

arXiv cs.CL Papers

Summary

This paper proposes HybridRAG-BN, a retrieval-augmented framework for Bangla knowledge-base question answering that combines hybrid retrieval, Gemma-based generation, and LoRA fine-tuned verification, achieving first place with F1 scores of 0.71654 and 0.72912.

arXiv:2608.13004v1 Announce Type: new Abstract: Knowledge-base question answering (KBQA) systems rely on effective retrieval and reasoning mechanisms to generate accurate answers from external knowledge sources. However, developing reliable KBQA systems for low-resource languages such as Bangla remains challenging due to limited retrieval-focused research, scarce language resources, and difficulties in grounding generated responses in external knowledge. In this work, we propose HybridRAG-BN, a retrieval-augmented framework for Bangla KBQA that integrates hybrid retrieval using BM25 and BGE-M3, answer generation using the GGUF version of Gemma-4-31B-Instruct, and a LoRA-fine-tuned Gemma-4-31B-Instruct model for answer verification and refinement. To further improve robustness, the framework incorporates a post-processing stage that addresses unresolved cases through fallback answer replacement and DuckDuckGo-assisted retrieval. Experimental results demonstrate the effectiveness of the proposed framework, achieving token-level F1 scores of 0.71654 and 0.72912 on the public and private leaderboards, respectively, securing first place in the competition.
Original Article
View Cached Full Text

Cached at: 08/14/26, 09:28 AM

# HybridRAG-BN: A Retrieval-Augmented Framework with Fine-Tuned Verification for Bangla KBQA
Source: [https://arxiv.org/abs/2608.13004](https://arxiv.org/abs/2608.13004)
[View PDF](https://arxiv.org/pdf/2608.13004)

> Abstract:Knowledge\-base question answering \(KBQA\) systems rely on effective retrieval and reasoning mechanisms to generate accurate answers from external knowledge sources\. However, developing reliable KBQA systems for low\-resource languages such as Bangla remains challenging due to limited retrieval\-focused research, scarce language resources, and difficulties in grounding generated responses in external knowledge\. In this work, we propose HybridRAG\-BN, a retrieval\-augmented framework for Bangla KBQA that integrates hybrid retrieval using BM25 and BGE\-M3, answer generation using the GGUF version of Gemma\-4\-31B\-Instruct, and a LoRA\-fine\-tuned Gemma\-4\-31B\-Instruct model for answer verification and refinement\. To further improve robustness, the framework incorporates a post\-processing stage that addresses unresolved cases through fallback answer replacement and DuckDuckGo\-assisted retrieval\. Experimental results demonstrate the effectiveness of the proposed framework, achieving token\-level F1 scores of 0\.71654 and 0\.72912 on the public and private leaderboards, respectively, securing first place in the competition\.

## Submission history

From: Nirjhar Das \[[view email](https://arxiv.org/show-email/c2c0315a/2608.13004)\] **\[v1\]**Thu, 13 Aug 2026 09:24:48 UTC \(534 KB\)

Similar Articles

When Retrieval Doesn't Help: A Large-Scale Study of Biomedical RAG

arXiv cs.CL

A large-scale study across 5 models (7B–72B), 10 biomedical QA datasets, 4 retrieval methods, and 4 corpora finds that RAG yields only small and inconsistent gains (1–2 points) over no-retrieval baselines in biomedical question answering. The study concludes that the main bottleneck is not retrieval quality but models' limited ability to effectively use retrieved evidence.