# Multi-session Conversation Question Answering on LoCoMo 98 QA probes Source: [Multi-session Conversation Question Answering on LoCoMo 98 QA probes]() Collection index: [llms.txt]() Data updated: 2026-04-15T08:35:55.677Z Evaluation of model performance on the LoCoMo 98 QA probes subset, focusing on multi-session conversational question answering capabilities. 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 | Multi-session Conversation Question Answering | | Dataset (including any stated split) | LoCoMo 98 QA probes | | Metric | Score (1-5) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 6 of 6 | | 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 - [Score (1-5)](): Higher is better; unit: SCALAR. - [95% CI Lower Bound (Score)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Score (1-5) | Result date | Report paper | | --- | --- | --- | --- | | Bookmark+Recall | 2.18 | Not recorded | 69df1b757d0172140ec7b7fb | | Full Context | 2.02 | Not recorded | 69df1b757d0172140ec7b7fb | | Search-Tool Baseline | 1.9 | Not recorded | 69df1b757d0172140ec7b7fb | | Word-Overlap Retrieval | 1.88 | Not recorded | 69df1b757d0172140ec7b7fb | | BM25 Retrieval | 1.86 | Not recorded | 69df1b757d0172140ec7b7fb | | Truncation | 1.64 | Not recorded | 69df1b757d0172140ec7b7fb | ### Bookmark+Recall Score (1-5): 2.18. Result ID: 69df4ddd47f34ce924c81688. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. ### Full Context Score (1-5): 2.02. Result ID: 69df4ddd47f34ce924c81687. [Reporting paper]() Recorded evaluation configuration: {"Context Window":"truncated to 60 turns","LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. ### Search-Tool Baseline Score (1-5): 1.9. Result ID: 69df4ddd47f34ce924c81686. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. ### Word-Overlap Retrieval Score (1-5): 1.88. Result ID: 69df4ddd47f34ce924c81685. [Reporting paper]() Recorded evaluation configuration: {"Retrieval":"top-3","LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. ### BM25 Retrieval Score (1-5): 1.86. Result ID: 69df4ddd47f34ce924c81684. [Reporting paper]() Recorded evaluation configuration: {"Retrieval":"top-3","LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. ### Truncation Score (1-5): 1.64. Result ID: 69df4ddd47f34ce924c81683. [Reporting paper]() Recorded evaluation configuration: {"Context Window":"20 turns","LLM":"GPT-4o-mini"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.