# Temporal Question Answering on TempReason OBQA-L2 Source: [Temporal Question Answering on TempReason OBQA-L2]() Collection index: [llms.txt]() Data updated: 2026-04-28T10:04:38.660Z Evaluation of model performance on the TempReason OBQA-L2 dataset for Temporal Question Answering, reporting Exact Match (EM) and F1 scores. 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 | Temporal Question Answering | | Dataset (including any stated split) | TempReason OBQA-L2 | | Metric | EM | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 - [EM](): Higher is better; unit: PERCENT. - [F1](): Higher is better; unit: PERCENT. ## Recorded results | Method | EM | Result date | Report paper | | --- | --- | --- | --- | | DeepSeek-V3-AdapTime | 48 | Not recorded | 69f03f6e55a3c32af89fa460 | | DeepSeek-V3-Step-back | 45.8 | Not recorded | 69f03f6e55a3c32af89fa460 | | GPT-4 | 45.4 | Not recorded | 69f03f6e55a3c32af89fa460 | | DeepSeek-V3-ICL | 45.1 | Not recorded | 69f03f6e55a3c32af89fa460 | | DeepSeek-V3-CoT | 44.8 | Not recorded | 69f03f6e55a3c32af89fa460 | | DeepSeek-V3-Self-refinement | 44.3 | Not recorded | 69f03f6e55a3c32af89fa460 | | TG-LLM | 42.4 | Not recorded | 69f03f6e55a3c32af89fa460 | | REMEMO-large | 37.4 | Not recorded | 69f03f6e55a3c32af89fa460 | | REMEMO-base | 33.6 | Not recorded | 69f03f6e55a3c32af89fa460 | | T5-large | 32.7 | Not recorded | 69f03f6e55a3c32af89fa460 | | Qwen-3-8B-AdapTime | 29.1 | Not recorded | 69f03f6e55a3c32af89fa460 | | T5-base | 26 | Not recorded | 69f03f6e55a3c32af89fa460 | | Qwen-3-8B-ICL | 23.9 | Not recorded | 69f03f6e55a3c32af89fa460 | | Qwen-3-8B-CoT | 22.6 | Not recorded | 69f03f6e55a3c32af89fa460 | | LLaMA-3-8B-AdapTime | 18.7 | Not recorded | 69f03f6e55a3c32af89fa460 | | LLaMA-3-8B-CoT | 18.5 | Not recorded | 69f03f6e55a3c32af89fa460 | | LLaMA-3-8B-ICL | 3.8 | Not recorded | 69f03f6e55a3c32af89fa460 | ### DeepSeek-V3-AdapTime EM: 48. Result ID: 69f0862447f34ce924d7007f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-V3-0324","Strategy":"AdapTime"}. This record does not assert independent reproduction. ### DeepSeek-V3-Step-back EM: 45.8. Result ID: 69f0862447f34ce924d7007d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-V3-0324","Strategy":"Step-back"}. This record does not assert independent reproduction. ### GPT-4 EM: 45.4. Result ID: 69f0862447f34ce924d70073. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"GPT-4","Strategy":"Vanilla"}. This record does not assert independent reproduction. ### DeepSeek-V3-ICL EM: 45.1. Result ID: 69f0862447f34ce924d7007b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-V3-0324","Strategy":"In-context Learning (ICL)"}. This record does not assert independent reproduction. ### DeepSeek-V3-CoT EM: 44.8. Result ID: 69f0862447f34ce924d7007c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-V3-0324","Strategy":"Chain-of-Thought (CoT)"}. This record does not assert independent reproduction. ### DeepSeek-V3-Self-refinement EM: 44.3. Result ID: 69f0862447f34ce924d7007e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-V3-0324","Strategy":"Self-refinement"}. This record does not assert independent reproduction. ### TG-LLM EM: 42.4. Result ID: 69f0862447f34ce924d70074. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Strategy":"Temporal Reasoning"}. This record does not assert independent reproduction. ### REMEMO-large EM: 37.4. Result ID: 69f0862447f34ce924d70072. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"REMEMO-large","Strategy":"Supervised Fine-tuning"}. This record does not assert independent reproduction. ### REMEMO-base EM: 33.6. Result ID: 69f0862447f34ce924d70071. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"REMEMO-base","Strategy":"Supervised Fine-tuning"}. This record does not assert independent reproduction. ### T5-large EM: 32.7. Result ID: 69f0862447f34ce924d70070. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"T5-large","Strategy":"Supervised Fine-tuning"}. This record does not assert independent reproduction. ### Qwen-3-8B-AdapTime EM: 29.1. Result ID: 69f0862447f34ce924d7007a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-3-8B","Strategy":"AdapTime"}. This record does not assert independent reproduction. ### T5-base EM: 26. Result ID: 69f0862447f34ce924d7006f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"T5-base","Strategy":"Supervised Fine-tuning"}. This record does not assert independent reproduction. ### Qwen-3-8B-ICL EM: 23.9. Result ID: 69f0862447f34ce924d70078. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen-3-8B","Strategy":"In-context Learning (ICL)"}. This record does not assert independent reproduction. ### Qwen-3-8B-CoT EM: 22.6. Result ID: 69f0862447f34ce924d70079. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen-3-8B","Strategy":"Chain-of-Thought (CoT)"}. This record does not assert independent reproduction. ### LLaMA-3-8B-AdapTime EM: 18.7. Result ID: 69f0862447f34ce924d70077. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaMA-3.1-8B-Instruct","Strategy":"AdapTime"}. This record does not assert independent reproduction. ### LLaMA-3-8B-CoT EM: 18.5. Result ID: 69f0862447f34ce924d70076. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LLaMA-3.1-8B-Instruct","Strategy":"Chain-of-Thought (CoT)"}. This record does not assert independent reproduction. ### LLaMA-3-8B-ICL EM: 3.8. Result ID: 69f0862447f34ce924d70075. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"LLaMA-3.1-8B-Instruct","Strategy":"In-context Learning (ICL)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.