# VAULT: Vigilant Adversarial Updates via LLM-Driven Retrieval-Augmented Generation for NLI

arXiv: https://arxiv.org/abs/2508.00965 · HTML: https://mlsanity.com/paper/2508.00965/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Privacy and security of RAG, one of 15 directions in Retrieval-augmented generation.
- Builds on 11 of its 15 references that are in the map.
- First posted on arXiv 1 Aug 2025.

## Placed in

- Retrieval-augmented generation / [Privacy and security of RAG](https://mlsanity.com/rag/privacy-and-security-of-rag/)

## Builds on (most cited first, 11 of 11)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [Adversarial Example Generation with Syntactically Controlled Paraphrase Networks](https://mlsanity.com/paper/1804.06059/) (2018)
- [Breaking NLI Systems with Sentences that Require Simple Lexical Inferences](https://mlsanity.com/paper/1805.02266/) (2018)
- [Adversarially Regularising Neural NLI Models to Integrate Logical Background Knowledge](https://mlsanity.com/paper/1808.08609/) (2018)
- [Prompting Large Language Models for Counterfactual Generation: An Empirical Study](https://mlsanity.com/paper/2305.14791/) (2023)
- [Behavior Analysis of NLI Models: Uncovering the Influence of Three Factors on Robustness](https://mlsanity.com/paper/1805.04212/) (2018)
- [A synthetic data approach for domain generalization of NLI models](https://mlsanity.com/paper/2402.12368/) (2024)
- [Don't Lag, RAG: Training-Free Adversarial Detection Using RAG](https://mlsanity.com/paper/2504.04858/) (2025)
- [Extracting and filtering paraphrases by bridging natural language inference and paraphrasing](https://mlsanity.com/paper/2111.07119/) (2021)

## Built on by (most cited first, 6 of 6)

- [FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents](https://mlsanity.com/paper/2606.12087/) (2026)
- [Seeing Isn't Believing: Context-Aware Adversarial Patch Synthesis via Conditional GAN](https://mlsanity.com/paper/2509.22836/) (2025)
- [NanoFlux: Adversarial Dual-LLM Evaluation and Distillation For Multi-Domain Reasoning](https://mlsanity.com/paper/2509.23252/) (2025)
- [RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-N Ranking](https://mlsanity.com/paper/2604.20623/) (2026)
- [Boundary on the Table: Efficient Black-Box Decision-Based Attacks for Structured Data](https://mlsanity.com/paper/2509.22850/) (2025)
- [Learning What to Fail On: Failure-Mode Contextual Bandits for Adversarial Data Curation](https://mlsanity.com/paper/2608.18681/) (2026)

## Abstract (arXiv)

We introduce VAULT, a fully automated adversarial RAG pipeline that systematically uncovers and remedies weaknesses in NLI models through three stages: retrieval, adversarial generation, and iterative retraining. First, we perform balanced few-shot retrieval by embedding premises with both semantic (BGE) and lexical (BM25) similarity. Next, we assemble these contexts into LLM prompts to generate adversarial hypotheses, which are then validated by an LLM ensemble for label fidelity. Finally, the validated adversarial examples are injected back into the training set at increasing mixing ratios, progressively fortifying a zero-shot RoBERTa-base model.On standard benchmarks, VAULT elevates RoBERTa-base accuracy from 88.48% to 92.60% on SNLI +4.12%, from 75.04% to 80.95% on ANLI +5.91%, and from 54.67% to 71.99% on MultiNLI +17.32%. It also consistently outperforms prior in-context adversarial methods by up to 2.0% across datasets. By automating high-quality adversarial data curation at scale, VAULT enables rapid, human-independent robustness improvements in NLI inference tasks.