# Scientific Reasoning on SCIENTIFIC Source: [Scientific Reasoning on SCIENTIFIC]() Collection index: [llms.txt]() Data updated: 2026-03-17T09:52:02.658Z Evaluation of model performance on the SCIENTIFIC dataset for Scientific Reasoning tasks, reporting Accuracy (Acc.) and Length (Len.). 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 | Scientific Reasoning | | Dataset (including any stated split) | SCIENTIFIC | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 36 of 36 | | 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. - [Length](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DEER | 82.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 80.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 80.3 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 77.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 74.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | ThinkLess | 74.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 74.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | NoThinking | 71.2 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 64.7 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 64.7 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 62.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 60.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Dynasor-CoT | 59.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Dynasor-CoT | 57.6 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 56.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 56.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | ThinkLess | 53.5 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 52 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 52 | Not recorded | 69b8e03b0636e5bf4afbed56 | | NoThinking | 51 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 50.5 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Dynasor-CoT | 48.5 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Dynasor-CoT | 45.5 | Not recorded | 69b8e03b0636e5bf4afbed56 | | NoThinking | 44.4 | Not recorded | 69b8e03b0636e5bf4afbed56 | | ThinkLess | 34.9 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 34.3 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 32.3 | Not recorded | 69b8e03b0636e5bf4afbed56 | | ThinkLess | 31.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | NoThinking | 31.3 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 28.3 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 25.3 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 24.8 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 21.2 | Not recorded | 69b8e03b0636e5bf4afbed56 | | RPDI-EE | 8.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | Vanilla CoT | 6.1 | Not recorded | 69b8e03b0636e5bf4afbed56 | | DEER | 4.6 | Not recorded | 69b8e03b0636e5bf4afbed56 | ### DEER Accuracy: 82.8. Result ID: 69b9243247f34ce924a241cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-235B-Thinking","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 80.8. Result ID: 69b9243247f34ce924a241cc. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-235B-Thinking","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 80.3. Result ID: 69b9243247f34ce924a241ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-235B-Thinking","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 77.8. Result ID: 69b9243247f34ce924a241c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-30B-A3B-Thinking-2507","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 74.8. Result ID: 69b9243247f34ce924a241bd. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-30B-A3B-Thinking-2507","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### ThinkLess Accuracy: 74.8. Result ID: 69b9243247f34ce924a241bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-30B-A3B-Thinking-2507","Reasoning Strategy":"ThinkLess"}. This record does not assert independent reproduction. ### DEER Accuracy: 74.8. Result ID: 69b9243247f34ce924a241c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-30B-A3B-Thinking-2507","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### NoThinking Accuracy: 71.2. Result ID: 69b9243247f34ce924a241be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-30B-A3B-Thinking-2507","Reasoning Strategy":"NoThinking"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 64.7. Result ID: 69b9243247f34ce924a241bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-32B","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 64.7. Result ID: 69b9243247f34ce924a241cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-70B","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### DEER Accuracy: 62.1. Result ID: 69b9243247f34ce924a241ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-70B","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 60.1. Result ID: 69b9243247f34ce924a241c9. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-70B","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### Dynasor-CoT Accuracy: 59.1. Result ID: 69b9243247f34ce924a241c0. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-30B-A3B-Thinking-2507","Reasoning Strategy":"Dynasor-CoT"}. This record does not assert independent reproduction. ### Dynasor-CoT Accuracy: 57.6. Result ID: 69b9243247f34ce924a241ba. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-32B","Reasoning Strategy":"Dynasor-CoT"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 56.1. Result ID: 69b9243247f34ce924a241b6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-14B","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 56.1. Result ID: 69b9243247f34ce924a241b7. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-32B","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### ThinkLess Accuracy: 53.5. Result ID: 69b9243247f34ce924a241b9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-32B","Reasoning Strategy":"ThinkLess"}. This record does not assert independent reproduction. ### DEER Accuracy: 52. Result ID: 69b9243247f34ce924a241b5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-14B","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### DEER Accuracy: 52. Result ID: 69b9243247f34ce924a241bb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-32B","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### NoThinking Accuracy: 51. Result ID: 69b9243247f34ce924a241b8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-32B","Reasoning Strategy":"NoThinking"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 50.5. Result ID: 69b9243247f34ce924a241b1. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-14B","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### Dynasor-CoT Accuracy: 48.5. Result ID: 69b9243247f34ce924a241b4. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-14B","Reasoning Strategy":"Dynasor-CoT"}. This record does not assert independent reproduction. ### Dynasor-CoT Accuracy: 45.5. Result ID: 69b9243247f34ce924a241ae. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Reasoning Strategy":"Dynasor-CoT"}. This record does not assert independent reproduction. ### NoThinking Accuracy: 44.4. Result ID: 69b9243247f34ce924a241b2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-14B","Reasoning Strategy":"NoThinking"}. This record does not assert independent reproduction. ### ThinkLess Accuracy: 34.9. Result ID: 69b9243247f34ce924a241b3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-14B","Reasoning Strategy":"ThinkLess"}. This record does not assert independent reproduction. ### DEER Accuracy: 34.3. Result ID: 69b9243247f34ce924a241c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-8B","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 32.3. Result ID: 69b9243247f34ce924a241ab. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### ThinkLess Accuracy: 31.8. Result ID: 69b9243247f34ce924a241ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Reasoning Strategy":"ThinkLess"}. This record does not assert independent reproduction. ### NoThinking Accuracy: 31.3. Result ID: 69b9243247f34ce924a241ac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Reasoning Strategy":"NoThinking"}. This record does not assert independent reproduction. ### DEER Accuracy: 28.3. Result ID: 69b9243247f34ce924a241af. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 25.3. Result ID: 69b9243247f34ce924a241c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-8B","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 24.8. Result ID: 69b9243247f34ce924a241b0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 21.2. Result ID: 69b9243247f34ce924a241c6. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Llama-8B","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### RPDI-EE Accuracy: 8.1. Result ID: 69b9243247f34ce924a241c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Reasoning Strategy":"RPDI-EE"}. This record does not assert independent reproduction. ### Vanilla CoT Accuracy: 6.1. Result ID: 69b9243247f34ce924a241c3. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Reasoning Strategy":"Vanilla CoT"}. This record does not assert independent reproduction. ### DEER Accuracy: 4.6. Result ID: 69b9243247f34ce924a241c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Reasoning Strategy":"DEER"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.