arXiv subjects

Sudhanshu Sharma

Publications and source records attributed to Sudhanshu Sharma.

4 recordsLinked to original sources

arXiv2026

Hybrid Retrieval-Augmented Generation with Knowledge Graph Expansion, RRF Fusion, and Per-Chunk Grounded Evaluation for Enterprise Document Search

Getting accurate, grounded answers out of large enterprise document repositories is a difficult problem. Dense vector retrieval alone frequently performs poorly on queries that mix technical terminology, vendor-specific acronyms, or require reasoning across several non-adjacent sections. DocuSearch was built to address exactly this gap - an offline, multi-agent document intelligence system developed and evaluated in a production telecom network operations environment. Rather than relying on a single retrieval signal, DocuSearch pulls together three complementary sources of evidence: semantic search over a Qdrant vector store using BGE-Large embeddings, BM25 full text search over an SQLite FTS5 index, and Knowledge Graph neighbour expansion from a structured edge table. These three ranked lists are merged through Reciprocal Rank Fusion with signal weights of 0.50 for vector search, 0.35 for BM25, and 0.15 for the knowledge graph, using a smoothing constant of 60 to stabilize scores. A cross-encoder then reranks the fused list, and Maximal Marginal Relevance with a balance factor of 0.65 prunes results for relevance and diversity. What makes DocuSearch distinctive is a per-chunk evaluation loop treating each chunk as its own mini-retrieval problem: an LLM decides whether the chunk needs more context, whether it fully answers the query, and whether the answer is grounded in retrieved text. Ungrounded answers are not returned; the system falls back to a multi-chunk merge instead. On a telecom corpus, DocuSearch reaches Precision@10 of 0.69, Recall@10 of 0.79, and a grounding rate of 89.6% - gains of 15, 16, and 18.4 percentage points over a dense-only RAG baseline. Index Terms: retrieval-augmented generation, knowledge graph, reciprocal rank fusion, enterprise document search, agentic evaluation, BM25, cross-encoder reranking, on-premise deployment, LangGraph, telecom AI.

cs.IR↗

arXiv2026

Multi-Agent Retrieval-Augmented Generation for Efficient Cloud Knowledge Base Search in Telecom SNOC Environment

Telecom Service and Network Operations Centers (SNOCs) rely on large collections of cloud documents, including Standard Operating Procedures (SOPs), vendor technical manuals, incident reports, and configuration guides, to maintain uninterrupted network operations. During critical incidents, engineers must quickly retrieve accurate information, yet traditional keyword based and single stage retrieval approaches often struggle to provide precise results. This paper presents Athena for Cloud Knowledge Base, a fully offline, multi agent Retrieval Augmented Generation (RAG) framework designed for enterprise cloud document search in Vodafone Idea's SNOC environment. The system integrates dense retrieval using E5 Large V2 embeddings, BM25 sparse retrieval, and Knowledge Graph expansion within a LangGraph based orchestration framework. Retrieved candidates are fused using Weighted CombSUM, followed by cross encoder reranking and Maximal Marginal Relevance (MMR) to obtain a diverse and relevant evidence set. To improve answer reliability, the framework performs per chunk LLM evaluation with explicit attribution verification, assessing each MMR selected chunk independently before generating a response. Unsupported or weak evidence is discarded, and if no chunk satisfies the verification criteria, the system automatically evaluates multiple chunks together as a fallback. Experiments on a corpus of 4200 SNOC cloud documents containing 312000 indexed chunks show that the proposed approach achieves an MRR at 10 of 0.910 and an Exact Match (EM) score of 78.4 percent, outperforming single stage dense retrieval by 14.6 percentage points. The entire pipeline operates in a fully offline environment, satisfying enterprise data sovereignty requirements while delivering accurate and grounded responses for cloud document search.

cs.IR↗

arXiv2022

High-yield exfoliation of MoS2 nanosheets by a novel spray technique and the importance of soaking and surfactants

Liquid-phase exfoliation of two-dimensional materials is very attractive for large-scale applications. Although used extensively, isolating MoS2 layers (<10) with high efficiency is reported to be extremely difficult. Further, the importance of soaking has not yet been studied, and the surfactants' role in stabilizing MoS2 nanosheets is poorly understood1. Herein, we report a novel approach to exfoliating large quantities of MoS2 via high-pressure (HP) liquid-phase exfoliation (LPE) in deionized (DI) water. 4 to 7 layers of MoS2 nanosheets were obtained from 60 days-soaked samples and they were found to be stable in solvents for periods of up to six months. Studies on the effect of three surfactants, namely sodium dodecyl benzenesulfonate (SDBS), sodium cholate (SC), and tetra-butyl ammonium bromide (TBAB), indicate that exfoliation of MoS2 nanosheets in SDBS is highly efficient than the other two surfactants. The estimated yield reaches up to 7.25%, with a nanosheet concentration of 1.45 mg/ml, which is one of the highest ever reported. Our studies also suggest that the nanosheets' concentration and the lateral size depend on exfoliation cycles, applied pressure and surfactant concentration. Hydrogen evolution reaction (HER) and ion-transport study show that the nanosheets prepared by our method are stable in an acidic medium and free from surfactants. A high hydrogen evolution rate of 30.13 mmol g-1 h-1 was estimated under ambient laboratory conditions.

cond-mat.mes-hall↗

arXiv2016

Autothermal Reforming of Methane on Rhodium Catalysts: Microkinetic Analysis for Model Reduction

Methane autothermal reforming has been studied using comprehensive, detailed microkinetic mechanisms, and a hierarchically reduced rate expression has been derived without apriori assumptions. The microkinetic mechanism is adapted from literature and has been validated with reported experimental results. Rate Determining Steps are elicited by reaction path analysis, partial equilibrium analysis and sensitivity analysis. Results show that methane activation occurs via dissociative adsorption to pyrolysis, while oxidation of the carbon occurs by O(s). Further, the mechanism is reduced through information obtained from the reaction path analysis, which is further substantiated by principal component analysis. A 33% reduction from the full microkinetic mechanism is obtained. One-step rate equation is further derived from the reduced microkinetic mechanism. The results show that the this rate equation accurately predicts conversions as well as outlet mole fraction for a wide range of inlet compositions.

cond-mat.mtrl-sci↗