# Abstention Enforcement on TruthfulQA no-context abstention stress test 100-item Source: [Abstention Enforcement on TruthfulQA no-context abstention stress test 100-item]() Collection index: [llms.txt]() Data updated: 2026-04-09T23:07:55.707Z Evaluation of model performance on the 100-item no-context abstention stress test subset of TruthfulQA, measuring correctness, hallucination, and abstention rates. 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 | Abstention Enforcement | | Dataset (including any stated split) | TruthfulQA no-context abstention stress test 100-item | | Metric | Correctness | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Correctness](): Higher is better; unit: PERCENT. - [Hallucination Rate](): Lower is better; unit: PERCENT. - [Abstention Rate](): Higher is better; unit: PERCENT. ## Recorded results | Method | Correctness | Result date | Report paper | | --- | --- | --- | --- | | GPT-4o-mini | 100 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o | 100 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o | 100 | Not recorded | 69d732617d0172140ec76957 | | GPT-3.5-turbo | 100 | Not recorded | 69d732617d0172140ec76957 | | GPT-3.5-turbo | 100 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o | 99 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o-mini | 98 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o-mini | 98 | Not recorded | 69d732617d0172140ec76957 | | GPT-3.5-turbo | 62 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o-mini | 0 | Not recorded | 69d732617d0172140ec76957 | | GPT-4o | 0 | Not recorded | 69d732617d0172140ec76957 | | GPT-3.5-turbo | 0 | Not recorded | 69d732617d0172140ec76957 | ### GPT-4o-mini Correctness: 100. Result ID: 69d8312a47f34ce924bfd079. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Instruction"}. This record does not assert independent reproduction. ### GPT-4o Correctness: 100. Result ID: 69d8312a47f34ce924bfd07d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Instruction"}. This record does not assert independent reproduction. ### GPT-4o Correctness: 100. Result ID: 69d8312a47f34ce924bfd07f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Composite"}. This record does not assert independent reproduction. ### GPT-3.5-turbo Correctness: 100. Result ID: 69d8312a47f34ce924bfd082. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Gate"}. This record does not assert independent reproduction. ### GPT-3.5-turbo Correctness: 100. Result ID: 69d8312a47f34ce924bfd083. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Composite"}. This record does not assert independent reproduction. ### GPT-4o Correctness: 99. Result ID: 69d8312a47f34ce924bfd07e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Gate"}. This record does not assert independent reproduction. ### GPT-4o-mini Correctness: 98. Result ID: 69d8312a47f34ce924bfd07a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Gate"}. This record does not assert independent reproduction. ### GPT-4o-mini Correctness: 98. Result ID: 69d8312a47f34ce924bfd07b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Composite"}. This record does not assert independent reproduction. ### GPT-3.5-turbo Correctness: 62. Result ID: 69d8312a47f34ce924bfd081. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Instruction"}. This record does not assert independent reproduction. ### GPT-4o-mini Correctness: 0. Result ID: 69d8312a47f34ce924bfd078. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Baseline"}. This record does not assert independent reproduction. ### GPT-4o Correctness: 0. Result ID: 69d8312a47f34ce924bfd07c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Baseline"}. This record does not assert independent reproduction. ### GPT-3.5-turbo Correctness: 0. Result ID: 69d8312a47f34ce924bfd080. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"Baseline"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.