# Conversational Question Answering on LoCoMo (test) Source: [Conversational Question Answering on LoCoMo (test)]() Collection index: [llms.txt]() Data updated: 2026-05-18T09:05:08.064Z Evaluation of model performance on the test split of the LoCoMo dataset for Conversational Question Answering, reporting F1 Score and Judge Score. 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 | Conversational Question Answering | | Dataset (including any stated split) | LoCoMo (test) | | Metric | F1 Score | | 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 - [F1 Score](): Higher is better; unit: PERCENT. - [Judge Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | F1 Score | Result date | Report paper | | --- | --- | --- | --- | | Full Context | 50.7 | Not recorded | 6a0a9d31dbc836b87d82b52b | | RecMem | 46.9 | Not recorded | 6a0a9d31dbc836b87d82b52b | | MemoryOS | 45.6 | Not recorded | 6a0a9d31dbc836b87d82b52b | | MemoryOS | 44.5 | Not recorded | 6a0a9d31dbc836b87d82b52b | | AMem | 44 | Not recorded | 6a0a9d31dbc836b87d82b52b | | Mem0 | 42.9 | Not recorded | 6a0a9d31dbc836b87d82b52b | | Mem0 | 42.8 | Not recorded | 6a0a9d31dbc836b87d82b52b | | Full Context | 42.3 | Not recorded | 6a0a9d31dbc836b87d82b52b | | RecMem | 38.5 | Not recorded | 6a0a9d31dbc836b87d82b52b | | AMem | 36.4 | Not recorded | 6a0a9d31dbc836b87d82b52b | | Naive Rag | 33.7 | Not recorded | 6a0a9d31dbc836b87d82b52b | | Naive Rag | 30.9 | Not recorded | 6a0a9d31dbc836b87d82b52b | ### Full Context F1 Score: 50.7. Result ID: 6a0ad63547f34ce924f12b6e. [Reporting paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4.1-MINI"}. This record does not assert independent reproduction. ### RecMem F1 Score: 46.9. Result ID: 6a0ad63547f34ce924f12b73. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4.1-MINI"}. This record does not assert independent reproduction. ### MemoryOS F1 Score: 45.6. Result ID: 6a0ad63547f34ce924f12b6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4O-MINI"}. This record does not assert independent reproduction. ### MemoryOS F1 Score: 44.5. Result ID: 6a0ad63547f34ce924f12b70. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4.1-MINI"}. This record does not assert independent reproduction. ### AMem F1 Score: 44. Result ID: 6a0ad63547f34ce924f12b72. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4.1-MINI"}. This record does not assert independent reproduction. ### Mem0 F1 Score: 42.9. Result ID: 6a0ad63547f34ce924f12b6b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4O-MINI"}. This record does not assert independent reproduction. ### Mem0 F1 Score: 42.8. Result ID: 6a0ad63547f34ce924f12b71. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4.1-MINI"}. This record does not assert independent reproduction. ### Full Context F1 Score: 42.3. Result ID: 6a0ad63547f34ce924f12b68. [Reporting paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4O-MINI"}. This record does not assert independent reproduction. ### RecMem F1 Score: 38.5. Result ID: 6a0ad63547f34ce924f12b6d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4O-MINI"}. This record does not assert independent reproduction. ### AMem F1 Score: 36.4. Result ID: 6a0ad63547f34ce924f12b6c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4O-MINI"}. This record does not assert independent reproduction. ### Naive Rag F1 Score: 33.7. Result ID: 6a0ad63547f34ce924f12b6f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4.1-MINI"}. This record does not assert independent reproduction. ### Naive Rag F1 Score: 30.9. Result ID: 6a0ad63547f34ce924f12b69. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM Backbone":"GPT-4O-MINI"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.