Jul 2026· Aadim Journal of Multidisciplinary Research· Vol 2, pp. 132-149· 0 citations· 28 references
TL;DR
Results suggest that code-mixing-aware RAG chatbots can meaningfully improve DSA learning outcomes while remaining feasible for instructor-led deployment in low-resource settings.
Abstract
Data structures and algorithms (DSA) form a cornerstone of computer science education, yet many students in Nepal struggle with the cognitive load imposed by English-only learning materials when their habitual conceptual language is a code-mixed register of Romanized Nepali and English. This paper presents the design, implementation, and classroom evaluation of an NLP-driven educational chatbot, AlgoSathi, that addresses this gap through a Retrieval-Augmented Generation (RAG) architecture tailored to code-mixed pedagogical queries. The system combines a curated DSA knowledge base of 1,252 chunks drawn from instructor-authored lecture notes, textbook material, and worked solutions; a multilingual sentence encoder (paraphrase-multilingual-mpnet-base-v2) backed by an HNSW-indexed FAISS vector store; a hybrid BM25-plus-dense retriever with a cross-encoder re-ranker; and a large language model generator constrained by a pedagogically structured prompt that preserves the language register of the learner's query. A 12-week quasi-experimental study with 52 undergraduates (26 experimental, 26 control) showed the chatbot group achieved a 26.2-point mean learning gain versus 14.5 points for controls (t (50) =4.21, p<0.001, d=1.17, 95% CI for gain difference: 6.2-17.2pp). ANCOVA using pre-test scores as a covariate confirmed a significant adjusted post-test advantage of 12.1 points (95% CI: 7.4–16.8, p < 0.001). Statistical assumption checks (Shapiro-Wilk, Levene's test) and a priori power analysis (power = 0.97) are reported. The retrieval pipeline reached P@5=0.79 on code-mixed queries, with student satisfaction averaging 4.31/5. Results suggest that code-mixing-aware RAG chatbots can meaningfully improve DSA learning outcomes while remaining feasible for instructor-led deployment in low-resource settings.
Universities now use Large Language Models (LLMs) to transform their processes for managing student information. The paper introduces an upgraded chatbot system for Narasaraopeta Engineering College (NEC) which extends previous on-premise LLM chatbot research by providing four new functions. The system uses (1) Retrieval-Augmented Generation (RAG) to create citation-based responses through LlamaIndex and ChromaDB, (2) Context Memory which maintains conversation flow during multiple dialogue exchanges, (3) Voice Input through OpenAI Whisper Speech-to-Text (STT) technology, and (4) Multilingual Support which covers English and these seven languages: Hindi, Telugu, Tamil, Kannada, and Malayalam through IndicNLP. The system tested 60 benchmark questions across four academic categories which included regulations and examination policies and fee structures and multilingual queries and achieved 96.7% overall accuracy with sub-second text response times and 1.0–1.4 second voice response times. The system operates entirely on-premise through Docker which safeguards institutional data privacy while eliminating the need for recurring cloud API expenses. The upcoming development will create Emotion-Aware AI, FAQ Auto-Learning, Student Portal Integration, and a Mobile Application.
M. Yaswanth, Kopparapu Sai Amar Durgesh, Mogili Harsha Vardhan et al.· 2026 7th International Confe...· 0 citations
It is concluded that an NLP- and RAG-based chatbot is feasible to be implemented as a digital campus information service and improves answer accuracy from 58.5% to 89.25%, with an average response time below three seconds.
Zuhri Yanda, Yeni Yanti, Maulinda et al.· IC-ITECHS· 0 citations
This paper presents the design and implementation of intelligent chatbots using Retrieval-Augmented Generation architecture to derive accurate, context-based responses from domain-specific knowledge sources such as PDFs and CSV files. The proposed system integrates information retrieval mechanisms with large language models to overcome crucial limitations of traditional chatbots: hallucination, knowledge cut-off, and inability to access private organizational data. The proposed architecture uses a six-stage pipeline consisting of document ingestion, text splitting, and embedding generation with sentence transformers, semantic retrieval, and response generation via Large Language Models (LLMs). By preprocessing the PDF and CSV documents using a RecursiveCharacterTextSplitter and converting them into 384-dimensional semantic embeddings, the system works via a similarity-based retrieval of document chunks. For implementation, the LlamaIndex framework was used for data indexing and query orchestration, and HuggingFace Transformers for language modeling. Experimental validation conducted on the smart campus chatbot system demonstrated significant improvements in the accuracy, relevance and transparency of responses across education, student services, and administrative support domains compared with a standalone language model. Because of its modular design, it supports a wide range of domain-specific topics without the combinatorial costs of re-training. This is another important application for systems that require regular updates and accurate information. Overall, the proposed framework provides a simple and practical approach for easily implementing domain-aware chatbots.
Varsha Spoorthi Shree Narravula, V. K, Yashaswini B. Suresh et al.· 2026 7th International Confe...· 0 citations
Foundation models, particularly Large Language Models (LLMs), show promise for chatbots for scholarly research chatbots but suffer from “hallucinations” and limited domain knowledge. This work addresses these reliability issues by developing a Retrieval-Augmented Generation (RAG) chatbot designed explicitly for scholarly research interactions within the Natural Language Processing (NLP) domain. The system utilizes a domain-specific corpus from the Association of Computational Linguistics (ACL) Anthology. The corpus was processed into text chunks and indexed using a high-dimensional vector database to facilitate efficient information retrieval. The generation component comprises an instruct-tuned LLM; Mixtral 8x7b, which synthesizes the retrieved context into coherent, evidence-based responses while maintaining the original semantic integrity of the scholarly source. Retrieval is performed using semantic similarity search. To improve dense retrieval precision, the BAAI/bge-large Sentence Transformer model was fine-tuned on a synthetic dataset generated from the corpus. Results show that fine-tuning and integrating the BAAI/bge-large model significantly improves the chatbot's ability to retrieve relevant information. Compared to the base model, retrieval accuracy of the fine-tuned model increases by up to 15% across different metrics: Accuracy@k, Precision@k, Recall@k, where k represents the count of retrieved chunks considered for evaluation, and Mean Reciprocal Rank (MRR@10). Notably, accuracy reaches an impressive 97%, demonstrating a significant boost in retrieving the most relevant scholarly information for user queries. These findings underscore the efficacy of fine-tuned Retrieval-Augmented Generation (RAG) systems in developing reliable, grounded chatbots for academic environments.
A. Oguntimilehin, Olamide Balogun· Nigerian Journal of Technolo...· 0 citations
Instruction tuning has become the standard method for adapting large language models to follow human intent, yet existing instruction datasets are dominated by English-language general-knowledge tasks and lack coverage of specialized pedagogical domains. This paper presents IKS-Instruct, a dataset of 24,795 instruction-response pairs for teaching language models to deliver educational content grounded in Indian Knowledge Systems (IKS). The dataset spans seven languages (English, Hindi, Sanskrit, Tamil, Telugu, Kannada, and Malayalam), covers 41 pedagogical techniques from the Vedic oral and mathematical traditions, and is aligned with the Central Board of Secondary Education (CBSE) curriculum for classes 6 through 12. The pairs are derived from six source types: classical text corpora (Bhagavad Gita, Thirukkural, Sangam literature, Vedic texts), curriculum-aligned pedagogical templates, Vedic mathematical sutra demonstrations, bilingual instruction pairs, technique-grounded multi-turn dialogues, and cross-tradition comparative analyses. Quality is assessed through a multi-judge evaluation framework in which independent language models score responses on 12 dimensions including technique fidelity, pedagogical quality, factual accuracy, and IKS cultural depth. Under a uniform five-judge external panel (median aggregation over 1,201 stratified items), the strongest IKS-Instruct fine-tune of a compact 7B model reaches a median judge score of 6.39, within 0.15 of a strong general-purpose reference model (Nemotron-Nano at 6.54) at a fraction of its deployment cost, while the base model without IKS fine-tuning scores near zero on the IKS-specific dimensions. Model quality does not increase monotonically with data curation, a result we report together with the corresponding data-quality gains.
Shwetha Singaravelu, G. Muruganantham, L. Rajendran et al.· 0 citations
The deployment of Large Language Models (LLMs) for low-resource languages is challenging due to the lack of linguistic resources, sparse digital content and the absence of structured knowledge bases. In this paper, we present an adaptive knowledge-augmented framework for Mizo Large Language Models by combining Retrieval-Augmented Generation (RAG) with continual learning. This methodology harnesses semantic retrieval with dense embeddings and FAISS indexing, adaptive evidence re-ranking, parameter-efficient fine-tuning, and incremental knowledge updating to enhance factual accuracy and decrease hallucinations. Experimental evaluation shows better retrieval performance, greater text creation quality, and superior human evaluation scores than typical multilingual LLMs and static RAG methods. Moreover, the continual learning technique allows for effective integration of newly accessible Mizo resources, without re-training the model from scratch. The suggested architecture offers a scalable, stable and reusable method for the development of intelligent language technologies for Mizo and other low-resource languages.
Vanlalropuia Ralte, Abhisake Sinha· International Journal For Mu...· 0 citations