EvidenT: Building Trustworthy Enterprise Assistants through Evidence Groundedness and Traceability

Anubha Kabra, Katie Jooyoung Kim, Colin Zhiwei Kou, Helene Sajer, Yimei Fan +3 more
9/18/2026
cs.AI
rag

No confident code implementation found for this paper.

Abstract

Enterprise AI assistants must produce responses that are verifiable and traceable to source evidence. However, retrieval augmented generation (RAG) over heterogeneous enterprise data can suffer from citation drift, unsupported content, and weak source traceability. We present EvidenT (T = Trust + Transparency + Traceability), a lightweight pipeline that verifies extracted evidence against retrieved documents before answer generation, without model retraining. EvidenT combines structured passage extraction with deterministic lexical alignment to filter unsupported content, correct citation drift, and preserve source-span traceability. On approximately 500 real enterprise queries, EvidenT improves gold-source hit rate by an average of 29% over prompting baselines, produces no citations to nonretrieved urls, and achieves near-saturated answer-to-source lexical coverage.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

CiteGuard-RAG: A Validation-Centered AI System for Evidence-Grounded Question Answering

84% match2026

RARE: Redundancy-Aware Retrieval Evaluation Framework for High-Similarity Corpora

82% match1 repo42026

Verified Misguidance: Measuring Structural Citation Failures in Search-Augmented LLMs

82% match2026

Beyond Self-Knowledge: Propagating Uncertainty Across Reasoning and Retrieval in LLMs

82% match2026

StateRAG: Typed State Contracts for Complex Retrieval-Augmented Generation

82% match2026

CiteShade: Citation Laundering in Multi-Source Retrieval-Augmented Generation and Its Counterfactual Defense

82% match2026

Cite this paper

@article{kabra2026evident,
  title  = {EvidenT: Building Trustworthy Enterprise Assistants through Evidence Groundedness and Traceability},
  author = {Anubha Kabra and Katie Jooyoung Kim and Colin Zhiwei Kou and Helene Sajer and Yimei Fan and Radomir Cisar and Heather Greenhalgh and Gabriel Martinez Vidiri},
  year   = {2026},
  eprint = {2609.22537},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2609.22537v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2609.22537.svg)](https://codeofpaper.com/paper/2609.22537)

Discussion