# Legal Contract Revision on CUAD Source: [Legal Contract Revision on CUAD]() Collection index: [llms.txt]() Data updated: 2026-04-14T10:11:51.072Z Evaluation of model performance on the Legal Contract Revision task using the CUAD dataset, measured by CQ, RRR, and TES scores. 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 | Legal Contract Revision | | Dataset (including any stated split) | CUAD | | Metric | CQ | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 25 of 25 | | 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 - [CQ](): Higher is better; unit: PERCENT. - [RRR](): Higher is better; unit: PERCENT. - [TES](): Higher is better; unit: PERCENT. ## Recorded results | Method | CQ | Result date | Report paper | | --- | --- | --- | --- | | RCBSF | 85.82 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RCBSF | 82.67 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RCBSF | 82.56 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RAG | 79.34 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RCBSF | 78.92 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Iteration | 78.12 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RAG | 77.56 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RAG | 76.45 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RCBSF | 76.22 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Iteration | 76.12 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Iteration | 75.23 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Iteration | 75.23 | Not recorded | 69ddca1e7d0172140ec79aa1 | | CoT | 75.23 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Standard | 72.89 | Not recorded | 69ddca1e7d0172140ec79aa1 | | CoT | 72.56 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RAG | 72.45 | Not recorded | 69ddca1e7d0172140ec79aa1 | | RAG | 72.45 | Not recorded | 69ddca1e7d0172140ec79aa1 | | CoT | 70.45 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Standard | 70.23 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Iteration | 70.12 | Not recorded | 69ddca1e7d0172140ec79aa1 | | CoT | 68.12 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Standard | 68.12 | Not recorded | 69ddca1e7d0172140ec79aa1 | | CoT | 67.56 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Standard | 66.78 | Not recorded | 69ddca1e7d0172140ec79aa1 | | Standard | 65.78 | Not recorded | 69ddca1e7d0172140ec79aa1 | ### RCBSF CQ: 85.82. Result ID: 69de12de47f34ce924c60f45. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Chat"}. This record does not assert independent reproduction. ### RCBSF CQ: 82.67. Result ID: 69de12de47f34ce924c60f31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### RCBSF CQ: 82.56. Result ID: 69de12de47f34ce924c60f3b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LawLLM-7B"}. This record does not assert independent reproduction. ### RAG CQ: 79.34. Result ID: 69de12de47f34ce924c60f43. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Chat"}. This record does not assert independent reproduction. ### RCBSF CQ: 78.92. Result ID: 69de12de47f34ce924c60f36. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral-7B"}. This record does not assert independent reproduction. ### Iteration CQ: 78.12. Result ID: 69de12de47f34ce924c60f44. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Chat"}. This record does not assert independent reproduction. ### RAG CQ: 77.56. Result ID: 69de12de47f34ce924c60f39. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LawLLM-7B"}. This record does not assert independent reproduction. ### RAG CQ: 76.45. Result ID: 69de12de47f34ce924c60f2f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### RCBSF CQ: 76.22. Result ID: 69de12de47f34ce924c60f40. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LexiLaw-6B"}. This record does not assert independent reproduction. ### Iteration CQ: 76.12. Result ID: 69de12de47f34ce924c60f3a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LawLLM-7B"}. This record does not assert independent reproduction. ### Iteration CQ: 75.23. Result ID: 69de12de47f34ce924c60f30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### Iteration CQ: 75.23. Result ID: 69de12de47f34ce924c60f35. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral-7B"}. This record does not assert independent reproduction. ### CoT CQ: 75.23. Result ID: 69de12de47f34ce924c60f42. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Chat"}. This record does not assert independent reproduction. ### Standard CQ: 72.89. Result ID: 69de12de47f34ce924c60f41. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Chat"}. This record does not assert independent reproduction. ### CoT CQ: 72.56. Result ID: 69de12de47f34ce924c60f2e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### RAG CQ: 72.45. Result ID: 69de12de47f34ce924c60f34. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Mistral-7B"}. This record does not assert independent reproduction. ### RAG CQ: 72.45. Result ID: 69de12de47f34ce924c60f3e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"LexiLaw-6B"}. This record does not assert independent reproduction. ### CoT CQ: 70.45. Result ID: 69de12de47f34ce924c60f38. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LawLLM-7B"}. This record does not assert independent reproduction. ### Standard CQ: 70.23. Result ID: 69de12de47f34ce924c60f2d. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B-Base"}. This record does not assert independent reproduction. ### Iteration CQ: 70.12. Result ID: 69de12de47f34ce924c60f3f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LexiLaw-6B"}. This record does not assert independent reproduction. ### CoT CQ: 68.12. Result ID: 69de12de47f34ce924c60f33. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Mistral-7B"}. This record does not assert independent reproduction. ### Standard CQ: 68.12. Result ID: 69de12de47f34ce924c60f37. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"LawLLM-7B"}. This record does not assert independent reproduction. ### CoT CQ: 67.56. Result ID: 69de12de47f34ce924c60f3d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"LexiLaw-6B"}. This record does not assert independent reproduction. ### Standard CQ: 66.78. Result ID: 69de12de47f34ce924c60f32. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Mistral-7B"}. This record does not assert independent reproduction. ### Standard CQ: 65.78. Result ID: 69de12de47f34ce924c60f3c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"LexiLaw-6B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.