# Confidence Estimation (Iterative Tagging) on WildHallu Source: [Confidence Estimation (Iterative Tagging) on WildHallu]() Collection index: [llms.txt]() Data updated: 2026-05-15T07:01:08.057Z Evaluation of model confidence estimation performance using iterative tagging methodology on the WildHallu dataset, measured using Brier Score (BS), ECE-M, and Scaled Calibration (SC). 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 | Confidence Estimation (Iterative Tagging) | | Dataset (including any stated split) | WildHallu | | Metric | Brier Score (BS) | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 9 of 9 | | 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 - [Brier Score (BS)](): Lower is better; unit: SCALAR. - [ECE-M](): Lower is better; unit: SCALAR. - [Scaled Calibration (SC)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Brier Score (BS) | Result date | Report paper | | --- | --- | --- | --- | | LOVEC-GRPO | 5.7 | Not recorded | 68fa78e61e209131cea7b12b | | LOVEC-DPO | 6 | Not recorded | 68fa78e61e209131cea7b12b | | LOVEC-SFT | 9.1 | Not recorded | 68fa78e61e209131cea7b12b | | Vanilla | 10.8 | Not recorded | 68fa78e61e209131cea7b12b | | LUQ | 14.5 | Not recorded | 68fa78e61e209131cea7b12b | | p(true)-ft | 16.4 | Not recorded | 68fa78e61e209131cea7b12b | | Self-Cons | 16.5 | Not recorded | 68fa78e61e209131cea7b12b | | Verb-Conf | 20.3 | Not recorded | 68fa78e61e209131cea7b12b | | p(true) | 23.8 | Not recorded | 68fa78e61e209131cea7b12b | ### LOVEC-GRPO Brier Score (BS): 5.7. Result ID: 6a06c4b047f34ce924edd306. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tagging Format":"Iterative","Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### LOVEC-DPO Brier Score (BS): 6. Result ID: 6a06c4b047f34ce924edd307. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tagging Format":"Iterative","Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### LOVEC-SFT Brier Score (BS): 9.1. Result ID: 6a06c4b047f34ce924edd305. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Tagging Format":"Iterative","Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### Vanilla Brier Score (BS): 10.8. Result ID: 6a06c4b047f34ce924edd300. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### LUQ Brier Score (BS): 14.5. Result ID: 6a06c4b047f34ce924edd2ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### p(true)-ft Brier Score (BS): 16.4. Result ID: 6a06c4b047f34ce924edd302. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### Self-Cons Brier Score (BS): 16.5. Result ID: 6a06c4b047f34ce924edd304. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### Verb-Conf Brier Score (BS): 20.3. Result ID: 6a06c4b047f34ce924edd303. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. ### p(true) Brier Score (BS): 23.8. Result ID: 6a06c4b047f34ce924edd301. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama3-8B-Instruct","Decoding Strategy":"Greedy"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.