# Commonsense Reasoning on LLM-adapters Reasoning (BoolQ, PIQA, SIQA, HellaSwag, ARC-c) Source: [Commonsense Reasoning on LLM-adapters Reasoning (BoolQ, PIQA, SIQA, HellaSwag, ARC-c)]() Collection index: [llms.txt]() Data updated: 2026-04-28T07:00:07.981Z Evaluation of model performance on a specific subset of commonsense reasoning tasks: BoolQ, PIQA, SIQA, HellaSwag, and ARC-c, reporting Accuracy and Memory usage. 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 | Commonsense Reasoning | | Dataset (including any stated split) | LLM-adapters Reasoning (BoolQ, PIQA, SIQA, HellaSwag, ARC-c) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher 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 - [Accuracy](): Higher is better; unit: PERCENT. - [Memory Usage](): Lower is better; unit: SIZE. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | LoRA | 76.19 | Not recorded | 69f03f0855a3c32af89f9f99 | | LARS-LP | 75.97 | Not recorded | 69f03f0855a3c32af89f9f99 | | AdaLoRA | 75.8 | Not recorded | 69f03f0855a3c32af89f9f99 | | LARS | 75.6 | Not recorded | 69f03f0855a3c32af89f9f99 | | IA3 | 71.66 | Not recorded | 69f03f0855a3c32af89f9f99 | | Prompt | 40.79 | Not recorded | 69f03f0855a3c32af89f9f99 | | LARS-LP | 39.32 | Not recorded | 69f03f0855a3c32af89f9f99 | | Prefix | 39.23 | Not recorded | 69f03f0855a3c32af89f9f99 | | AdaLoRA | 38.4 | Not recorded | 69f03f0855a3c32af89f9f99 | | IA3 | 37.44 | Not recorded | 69f03f0855a3c32af89f9f99 | | LARS | 37.33 | Not recorded | 69f03f0855a3c32af89f9f99 | | LoRA | 35.73 | Not recorded | 69f03f0855a3c32af89f9f99 | | Prompt | 15.47 | Not recorded | 69f03f0855a3c32af89f9f99 | | Prefix | 9.03 | Not recorded | 69f03f0855a3c32af89f9f99 | ### LoRA Accuracy: 76.19. Result ID: 69f05a5547f34ce924d5a5e6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.45"}. This record does not assert independent reproduction. ### LARS-LP Accuracy: 75.97. Result ID: 69f05a5547f34ce924d5a5ec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.67"}. This record does not assert independent reproduction. ### AdaLoRA Accuracy: 75.8. Result ID: 69f05a5547f34ce924d5a5e7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.45"}. This record does not assert independent reproduction. ### LARS Accuracy: 75.6. Result ID: 69f05a5547f34ce924d5a5eb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.67"}. This record does not assert independent reproduction. ### IA3 Accuracy: 71.66. Result ID: 69f05a5547f34ce924d5a5e8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.01"}. This record does not assert independent reproduction. ### Prompt Accuracy: 40.79. Result ID: 69f05a5547f34ce924d5a5ea. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.003"}. This record does not assert independent reproduction. ### LARS-LP Accuracy: 39.32. Result ID: 69f05a5547f34ce924d5a5f3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.67"}. This record does not assert independent reproduction. ### Prefix Accuracy: 39.23. Result ID: 69f05a5547f34ce924d5a5e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama-3.2-1B","Trainable Parameters (%)":"0.03"}. This record does not assert independent reproduction. ### AdaLoRA Accuracy: 38.4. Result ID: 69f05a5547f34ce924d5a5ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.45"}. This record does not assert independent reproduction. ### IA3 Accuracy: 37.44. Result ID: 69f05a5547f34ce924d5a5ef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.01"}. This record does not assert independent reproduction. ### LARS Accuracy: 37.33. Result ID: 69f05a5547f34ce924d5a5f2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.67"}. This record does not assert independent reproduction. ### LoRA Accuracy: 35.73. Result ID: 69f05a5547f34ce924d5a5ed. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.45"}. This record does not assert independent reproduction. ### Prompt Accuracy: 15.47. Result ID: 69f05a5547f34ce924d5a5f1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.003"}. This record does not assert independent reproduction. ### Prefix Accuracy: 9.03. Result ID: 69f05a5547f34ce924d5a5f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Instruct","Trainable Parameters (%)":"0.03"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.