# Extractive Question Answering for Causal Relations on FinCausal Spanish 2026 (test) Source: [Extractive Question Answering for Causal Relations on FinCausal Spanish 2026 (test)]() Collection index: [llms.txt]() Data updated: 2026-06-29T06:26:09.156Z Evaluation of model performance on the test split of the FinCausal Spanish 2026 dataset for Extractive Question Answering focused on Causal Relations. 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 | Extractive Question Answering for Causal Relations | | Dataset (including any stated split) | FinCausal Spanish 2026 (test) | | Metric | LLM Judge Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 - [LLM Judge Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | LLM Judge Score | Result date | Report paper | | --- | --- | --- | --- | | HSA\_CORAL | 4.7753 | Not recorded | 6a41fc11206542251d7ccae3 | | GPT-4.1 mini | 4.7674 | Not recorded | 6a41fc11206542251d7ccae3 | | GPT-5.2 | 4.735 | Not recorded | 6a41fc11206542251d7ccae3 | | GPT-4.1 mini | 4.7141 | Not recorded | 6a41fc11206542251d7ccae3 | | GPT-3.5 turbo | 4.706 | Not recorded | 6a41fc11206542251d7ccae3 | | BART Facebook Base | 4.03 | Not recorded | 6a41fc11206542251d7ccae3 | | BERT Base Multilingual | 3.981 | Not recorded | 6a41fc11206542251d7ccae3 | | Llama-3.1 8B | 3.91 | Not recorded | 6a41fc11206542251d7ccae3 | ### HSA\_CORAL LLM Judge Score: 4.7753. Result ID: 6a420ffafbc373592c865202. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"GPT-4.1 mini","Fine-tuned":"yes","Corpus":"Multilingual (en+es)"}. This record does not assert independent reproduction. ### GPT-4.1 mini LLM Judge Score: 4.7674. Result ID: 6a420ffafbc373592c865201. [Reporting paper]() Recorded evaluation configuration: {"Fine-tuned":"yes","Corpus":"Monolingual (es)"}. This record does not assert independent reproduction. ### GPT-5.2 LLM Judge Score: 4.735. Result ID: 6a420ffafbc373592c865203. [Reporting paper]() Recorded evaluation configuration: {"Fine-tuned":"no","Corpus":"-"}. This record does not assert independent reproduction. ### GPT-4.1 mini LLM Judge Score: 4.7141. Result ID: 6a420ffafbc373592c865200. [Reporting paper]() Recorded evaluation configuration: {"Fine-tuned":"yes","Corpus":"Monolingual (en)"}. This record does not assert independent reproduction. ### GPT-3.5 turbo LLM Judge Score: 4.706. Result ID: 6a420ffafbc373592c8651ff. [Reporting paper]() Recorded evaluation configuration: {"Fine-tuned":"no","Corpus":"-"}. This record does not assert independent reproduction. ### BART Facebook Base LLM Judge Score: 4.03. Result ID: 6a420ffafbc373592c8651fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Fine-tuned":"yes","Corpus":"Multilingual (en+es)"}. This record does not assert independent reproduction. ### BERT Base Multilingual LLM Judge Score: 3.981. Result ID: 6a420ffafbc373592c8651fc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Fine-tuned":"yes","Corpus":"Multilingual (en+es)"}. This record does not assert independent reproduction. ### Llama-3.1 8B LLM Judge Score: 3.91. Result ID: 6a420ffafbc373592c8651fe. [Reporting paper]() Recorded evaluation configuration: {"Fine-tuned":"yes","Corpus":"Multilingual (en+es)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.