# Question Answering on NQ PIA Attack (test) Source: [Question Answering on NQ PIA Attack (test)]() Collection index: [llms.txt]() Data updated: 2026-05-22T08:10:46.538Z Evaluation of model performance on the test split of the NQ PIA Attack dataset for Question Answering, measured by Accuracy (Acc.) and Answer Success Rate (ASR). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Question Answering | | Dataset (including any stated split) | NQ PIA Attack (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. - [Answer Success Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | RobustRAG | 66.2 | Not recorded | 6a0fe329f984739ed6739ec5 | | ReliabilityRAG | 66.2 | Not recorded | 6a0fe329f984739ed6739ec5 | | RADAR | 66.2 | Not recorded | 6a0fe329f984739ed6739ec5 | | RobustRAG | 65.4 | Not recorded | 6a0fe329f984739ed6739ec5 | | RADAR | 65.4 | Not recorded | 6a0fe329f984739ed6739ec5 | | RADAR | 65 | Not recorded | 6a0fe329f984739ed6739ec5 | | RobustRAG | 62 | Not recorded | 6a0fe329f984739ed6739ec5 | | InstructRAG | 59.6 | Not recorded | 6a0fe329f984739ed6739ec5 | | InstructRAG | 58.6 | Not recorded | 6a0fe329f984739ed6739ec5 | | ReliabilityRAG | 56.6 | Not recorded | 6a0fe329f984739ed6739ec5 | | ReliabilityRAG | 51.2 | Not recorded | 6a0fe329f984739ed6739ec5 | | Vanilla RAG | 47.6 | Not recorded | 6a0fe329f984739ed6739ec5 | | Vanilla RAG | 41 | Not recorded | 6a0fe329f984739ed6739ec5 | | AstuteRAG | 29 | Not recorded | 6a0fe329f984739ed6739ec5 | | InstructRAG | 27.8 | Not recorded | 6a0fe329f984739ed6739ec5 | | AstuteRAG | 27.4 | Not recorded | 6a0fe329f984739ed6739ec5 | | AstuteRAG | 22 | Not recorded | 6a0fe329f984739ed6739ec5 | | Vanilla RAG | 15.2 | Not recorded | 6a0fe329f984739ed6739ec5 | ### RobustRAG Accuracy: 66.2. Result ID: 6a100f82fbc373592c08e38c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"50","top-k":"50"}. This record does not assert independent reproduction. ### ReliabilityRAG Accuracy: 66.2. Result ID: 6a100f82fbc373592c08e38f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"50","top-k":"50"}. This record does not assert independent reproduction. ### RADAR Accuracy: 66.2. Result ID: 6a100f82fbc373592c08e392. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"50","top-k":"50"}. This record does not assert independent reproduction. ### RobustRAG Accuracy: 65.4. Result ID: 6a100f82fbc373592c08e38b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"25","top-k":"50"}. This record does not assert independent reproduction. ### RADAR Accuracy: 65.4. Result ID: 6a100f82fbc373592c08e391. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"25","top-k":"50"}. This record does not assert independent reproduction. ### RADAR Accuracy: 65. Result ID: 6a100f82fbc373592c08e390. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"1","top-k":"50"}. This record does not assert independent reproduction. ### RobustRAG Accuracy: 62. Result ID: 6a100f82fbc373592c08e38a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"1","top-k":"50"}. This record does not assert independent reproduction. ### InstructRAG Accuracy: 59.6. Result ID: 6a100f82fbc373592c08e389. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"50","top-k":"50"}. This record does not assert independent reproduction. ### InstructRAG Accuracy: 58.6. Result ID: 6a100f82fbc373592c08e388. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"25","top-k":"50"}. This record does not assert independent reproduction. ### ReliabilityRAG Accuracy: 56.6. Result ID: 6a100f82fbc373592c08e38e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"25","top-k":"50"}. This record does not assert independent reproduction. ### ReliabilityRAG Accuracy: 51.2. Result ID: 6a100f82fbc373592c08e38d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"1","top-k":"50"}. This record does not assert independent reproduction. ### Vanilla RAG Accuracy: 47.6. Result ID: 6a100f82fbc373592c08e382. [Reporting paper]() Recorded evaluation configuration: {"Pos":"25","top-k":"50"}. This record does not assert independent reproduction. ### Vanilla RAG Accuracy: 41. Result ID: 6a100f82fbc373592c08e383. [Reporting paper]() Recorded evaluation configuration: {"Pos":"50","top-k":"50"}. This record does not assert independent reproduction. ### AstuteRAG Accuracy: 29. Result ID: 6a100f82fbc373592c08e385. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"25","top-k":"50"}. This record does not assert independent reproduction. ### InstructRAG Accuracy: 27.8. Result ID: 6a100f82fbc373592c08e387. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"1","top-k":"50"}. This record does not assert independent reproduction. ### AstuteRAG Accuracy: 27.4. Result ID: 6a100f82fbc373592c08e386. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"50","top-k":"50"}. This record does not assert independent reproduction. ### AstuteRAG Accuracy: 22. Result ID: 6a100f82fbc373592c08e384. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pos":"1","top-k":"50"}. This record does not assert independent reproduction. ### Vanilla RAG Accuracy: 15.2. Result ID: 6a100f82fbc373592c08e381. [Reporting paper]() Recorded evaluation configuration: {"Pos":"1","top-k":"50"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.