# Streaming Long-context Question Answering on STEM-Bench The Big Bang Theory Source: [Streaming Long-context Question Answering on STEM-Bench The Big Bang Theory]() Collection index: [llms.txt]() Data updated: 2026-03-06T08:11:00.074Z Evaluation of model performance on the STEM-Bench The Big Bang Theory dataset specifically designed for streaming long-context question answering tasks. 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 | Streaming Long-context Question Answering | | Dataset (including any stated split) | STEM-Bench The Big Bang Theory | | Metric | Time (s) | | Unit | TIME | | Direction | Lower is better | | Results loaded | 14 of 14 | | 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 - [Time (s)](): Lower is better; unit: TIME. - [B-4 Score](): Higher is better; unit: SCALAR. - [R-L Score](): Higher is better; unit: SCALAR. - [B-S Score](): Higher is better; unit: SCALAR. - [Evidence Score](): Higher is better; unit: PERCENT. - [KEM Score](): Higher is better; unit: SCALAR. - [Gem Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Time (s) | Result date | Report paper | | --- | --- | --- | --- | | ProStream (-w/o Mtree) | 0.4 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | ProStream (-w/o Mpend) | 0.401 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | ProStream | 0.408 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | ProStream (-w/o B) | 0.409 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | RAPTOR | 0.519 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | RAG | 0.957 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | GraphRAG | 0.997 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | A-Mem | 1.301 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | Summary | 2.001 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | HippoRAG2 | 4.231 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | RQ-RAG | 5.149 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | MemGAS | 5.711 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | Full Context | 7.333 | Not recorded | 69aa5f6e2d603e6de3e417d1 | | MemoRAG | 7.498 | Not recorded | 69aa5f6e2d603e6de3e417d1 | ### ProStream (-w/o Mtree) Time (s): 0.4. Result ID: 69aa8c116dbb7075b95430f4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"ablation":"Removing the Hierarchical Tree"}. This record does not assert independent reproduction. ### ProStream (-w/o Mpend) Time (s): 0.401. Result ID: 69aa8c116dbb7075b95430f5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"ablation":"Eliminating Pending buffer"}. This record does not assert independent reproduction. ### ProStream Time (s): 0.408. Result ID: 69aa8c116dbb7075b95430f2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ProStream (-w/o B) Time (s): 0.409. Result ID: 69aa8c116dbb7075b95430f3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"ablation":"Eliminating the Short-Term Sensing Buffer"}. This record does not assert independent reproduction. ### RAPTOR Time (s): 0.519. Result ID: 69aa8c116dbb7075b95430ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RAG Time (s): 0.957. Result ID: 69aa8c116dbb7075b95430e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GraphRAG Time (s): 0.997. Result ID: 69aa8c116dbb7075b95430ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### A-Mem Time (s): 1.301. Result ID: 69aa8c116dbb7075b95430f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Summary Time (s): 2.001. Result ID: 69aa8c116dbb7075b95430ec. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### HippoRAG2 Time (s): 4.231. Result ID: 69aa8c116dbb7075b95430ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RQ-RAG Time (s): 5.149. Result ID: 69aa8c116dbb7075b95430e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MemGAS Time (s): 5.711. Result ID: 69aa8c116dbb7075b95430f1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Full Context Time (s): 7.333. Result ID: 69aa8c116dbb7075b95430eb. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MemoRAG Time (s): 7.498. Result ID: 69aa8c116dbb7075b95430ef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.