# Spoken Language Understanding on Internal Traffic Dataset (regular test) Source: [Spoken Language Understanding on Internal Traffic Dataset (regular test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:46:47.879Z Evaluation of Spoken Language Understanding performance on the regular test split of the Internal Traffic Dataset, measured using ICER, SER, and IRER metrics. 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 | Spoken Language Understanding | | Dataset (including any stated split) | Internal Traffic Dataset (regular test) | | Metric | ICER Improvement (%) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 5 of 5 | | 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 - [ICER Improvement (%)](): Higher is better; unit: PERCENT. - [SER Improvement (%)](): Higher is better; unit: PERCENT. - [IRER Improvement (%)](): Higher is better; unit: PERCENT. ## Recorded results | Method | ICER Improvement (%) | Result date | Report paper | | --- | --- | --- | --- | | transformer-based E2E SLU | 9.3 | Not recorded | 68fa6aef1e209131cea2b2d7 | | transformer-based E2E SLU | 9.1 | Not recorded | 68fa6aef1e209131cea2b2d7 | | Multitask Transformer Baseline | 4 | Not recorded | 68fa6aef1e209131cea2b2d7 | | Multitask BiLSTM Baseline | 0.5 | Not recorded | 68fa6aef1e209131cea2b2d7 | | Multitask LSTM Baseline | 0 | Not recorded | 68fa6aef1e209131cea2b2d7 | ### transformer-based E2E SLU ICER Improvement (%): 9.3. Result ID: 698e4365ffc3901de09c2189. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-training strategy":"BERT, AC Pretraining","Num. Params.":"115M"}. This record does not assert independent reproduction. ### transformer-based E2E SLU ICER Improvement (%): 9.1. Result ID: 698e4365ffc3901de09c2188. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-training strategy":"No Pretraining","Num. Params.":"115M"}. This record does not assert independent reproduction. ### Multitask Transformer Baseline ICER Improvement (%): 4. Result ID: 698e4365ffc3901de09c2187. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Num. Params.":"101M"}. This record does not assert independent reproduction. ### Multitask BiLSTM Baseline ICER Improvement (%): 0.5. Result ID: 698e4365ffc3901de09c2186. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Num. Params.":"67.3M"}. This record does not assert independent reproduction. ### Multitask LSTM Baseline ICER Improvement (%): 0. Result ID: 698e4365ffc3901de09c2185. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Num. Params.":"45.9M"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.