# Evidence Contextualization and Counterfactual Attribution for Conversational QA over Heterogeneous Data with RAG Systems

arXiv: https://arxiv.org/abs/2412.10571 · HTML: https://mlsanity.com/paper/2412.10571/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation.
- Builds on 10 of its 17 references that are in the map.
- First posted on arXiv 13 Dec 2024.

## Placed in

- Retrieval-augmented generation / [RAG robustness and adaptive retrieval](https://mlsanity.com/rag/rag-robustness-and-adaptive-retrieval/)

## Builds on (most cited first, 10 of 10)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [CoQA: A Conversational Question Answering Challenge](https://mlsanity.com/paper/1808.07042/) (2018)
- [Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts](https://mlsanity.com/paper/2305.13300/) (2023)
- [Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding](https://mlsanity.com/paper/2401.04398/) (2024)
- [A Survey of Large Language Models Attribution](https://mlsanity.com/paper/2311.03731/) (2023)
- [Table-GPT: Table-tuned GPT for Diverse Table Tasks](https://mlsanity.com/paper/2310.09263/) (2023)
- [UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering](https://mlsanity.com/paper/2012.14610/) (2020)
- [Look before you Hop: Conversational Question Answering over Knowledge Graphs Using Judicious Context Expansion](https://mlsanity.com/paper/1910.03262/) (2019)
- [Counterfactual Explanations for Neural Recommenders](https://mlsanity.com/paper/2105.05008/) (2021)

## Built on by (most cited first, 8 of 8)

- [Attribution, Citation, and Quotation: A Survey of Evidence-based Text Generation with Large Language Models](https://mlsanity.com/paper/2508.15396/) (2025)
- [Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents](https://mlsanity.com/paper/2605.17830/) (2026)
- [RAGXplain: From Explainable Evaluation to Actionable Guidance of RAG Pipelines](https://mlsanity.com/paper/2505.13538/) (2025)
- [CUE-R: Beyond the Final Answer in Retrieval-Augmented Generation](https://mlsanity.com/paper/2604.05467/) (2026)
- [Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation](https://mlsanity.com/paper/2601.00263/) (2026)
- [Comprehensive Comparison of RAG Methods Across Multi-Domain Conversational QA](https://mlsanity.com/paper/2602.09552/) (2026)
- [What Would Fix This RAG Failure? Auditing Counterfactual Response with Paired Evidence Interventions](https://mlsanity.com/paper/2608.08944/) (2026)
- [MeClear: Cooperative Game-Theoretic Attribution and Risk-Aware Memory Clearance for Long-Horizon LLM Agents](https://mlsanity.com/paper/2609.09115/) (2026)

## Abstract (arXiv)

Retrieval Augmented Generation (RAG) works as a backbone for interacting with an enterprise's own data via Conversational Question Answering (ConvQA). In a RAG system, a retriever fetches passages from a collection in response to a question, which are then included in the prompt of a large language model (LLM) for generating a natural language (NL) answer. However, several RAG systems today suffer from two shortcomings: (i) retrieved passages usually contain their raw text and lack appropriate document context, negatively impacting both retrieval and answering quality; and (ii) attribution strategies that explain answer generation typically rely only on similarity between the answer and the retrieved passages, thereby only generating plausible but not causal explanations. In this work, we demonstrate RAGONITE, a RAG system that remedies the above concerns by: (i) contextualizing evidence with source metadata and surrounding text; and (ii) computing counterfactual attribution, a causal explanation approach where the contribution of an evidence to an answer is determined by the similarity of the original response to the answer obtained by removing that evidence. To evaluate our proposals, we release a new benchmark ConfQuestions: it has 300 hand-created conversational questions, each in English and German, coupled with ground truth URLs, completed questions, and answers from 215 public Confluence pages. These documents are typical of enterprise wiki spaces with heterogeneous elements. Experiments with RAGONITE on ConfQuestions show the viability of our ideas: contextualization improves RAG performance, and counterfactual explanations outperform standard attribution.