# Spreadsheet Understanding on SpreadsheetLLM benchmark Source: [Spreadsheet Understanding on SpreadsheetLLM benchmark]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:56.318Z Evaluation of model performance on the Spreadsheet Understanding task using the SpreadsheetLLM benchmark dataset, typically reporting Answer Accuracy and Latency. 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 | Spreadsheet Understanding | | Dataset (including any stated split) | SpreadsheetLLM benchmark | | Metric | Answer Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Answer Accuracy](): Higher is better; unit: PERCENT. - [Latency (s)](): Lower is better; unit: TIME. ## Recorded results | Method | Answer Accuracy | Result date | Report paper | | --- | --- | --- | --- | | SpreadsheetLLM | 91 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | SpreadsheetLLM | 90 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | SpreadsheetLLM | 89 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | FRTR | 87 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | FRTR | 85 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | FRTR | 84 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | SpreadsheetLLM | 80 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | SpreadsheetLLM | 78 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | FRTR | 77 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | FRTR | 75 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | SpreadsheetLLM | 73 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | Naïve Full-Context | 68 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | Naïve Full-Context | 67 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | Naïve Full-Context | 59 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | Naïve Full-Context | 55 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | FRTR | 53 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | Naïve Full-Context | 41 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | | Naïve Full-Context | 35 | Not recorded | 6967232ce25ad2fa9d1ec2b5 | ### SpreadsheetLLM Answer Accuracy: 91. Result ID: 699688795698665ec4d0f1e4. [Reporting paper]() Recorded evaluation configuration: {"Technique":"SpreadsheetLLM","Model":"Claude Sonnet 4.5"}. This record does not assert independent reproduction. ### SpreadsheetLLM Answer Accuracy: 90. Result ID: 699688795698665ec4d0f1e3. [Reporting paper]() Recorded evaluation configuration: {"Technique":"SpreadsheetLLM","Model":"GPT-5"}. This record does not assert independent reproduction. ### SpreadsheetLLM Answer Accuracy: 89. Result ID: 699688795698665ec4d0f1e5. [Reporting paper]() Recorded evaluation configuration: {"Technique":"SpreadsheetLLM","Model":"Gemini 2.5 Pro"}. This record does not assert independent reproduction. ### FRTR Answer Accuracy: 87. Result ID: 699688795698665ec4d0f1dd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Technique":"FRTR","Model":"GPT-5"}. This record does not assert independent reproduction. ### FRTR Answer Accuracy: 85. Result ID: 699688795698665ec4d0f1df. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Technique":"FRTR","Model":"Gemini 2.5 Pro"}. This record does not assert independent reproduction. ### FRTR Answer Accuracy: 84. Result ID: 699688795698665ec4d0f1de. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Technique":"FRTR","Model":"Claude Sonnet 4.5"}. This record does not assert independent reproduction. ### SpreadsheetLLM Answer Accuracy: 80. Result ID: 699688795698665ec4d0f1e7. [Reporting paper]() Recorded evaluation configuration: {"Technique":"SpreadsheetLLM","Model":"GPT-OSS-120b"}. This record does not assert independent reproduction. ### SpreadsheetLLM Answer Accuracy: 78. Result ID: 699688795698665ec4d0f1e2. [Reporting paper]() Recorded evaluation configuration: {"Technique":"SpreadsheetLLM","Model":"GPT-4o"}. This record does not assert independent reproduction. ### FRTR Answer Accuracy: 77. Result ID: 699688795698665ec4d0f1dc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Technique":"FRTR","Model":"GPT-4o"}. This record does not assert independent reproduction. ### FRTR Answer Accuracy: 75. Result ID: 699688795698665ec4d0f1e1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Technique":"FRTR","Model":"GPT-OSS-120b"}. This record does not assert independent reproduction. ### SpreadsheetLLM Answer Accuracy: 73. Result ID: 699688795698665ec4d0f1e6. [Reporting paper]() Recorded evaluation configuration: {"Technique":"SpreadsheetLLM","Model":"LLaMA Maverick-17B-instruct"}. This record does not assert independent reproduction. ### Naïve Full-Context Answer Accuracy: 68. Result ID: 699688795698665ec4d0f1e9. [Reporting paper]() Recorded evaluation configuration: {"Technique":"Naïve Full-Context","Model":"GPT-5"}. This record does not assert independent reproduction. ### Naïve Full-Context Answer Accuracy: 67. Result ID: 699688795698665ec4d0f1eb. [Reporting paper]() Recorded evaluation configuration: {"Technique":"Naïve Full-Context","Model":"Gemini 2.5 Pro"}. This record does not assert independent reproduction. ### Naïve Full-Context Answer Accuracy: 59. Result ID: 699688795698665ec4d0f1ed. [Reporting paper]() Recorded evaluation configuration: {"Technique":"Naïve Full-Context","Model":"GPT-OSS-120b"}. This record does not assert independent reproduction. ### Naïve Full-Context Answer Accuracy: 55. Result ID: 699688795698665ec4d0f1ea. [Reporting paper]() Recorded evaluation configuration: {"Technique":"Naïve Full-Context","Model":"Claude Sonnet 4.5"}. This record does not assert independent reproduction. ### FRTR Answer Accuracy: 53. Result ID: 699688795698665ec4d0f1e0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Technique":"FRTR","Model":"LLaMA Maverick-17B-instruct"}. This record does not assert independent reproduction. ### Naïve Full-Context Answer Accuracy: 41. Result ID: 699688795698665ec4d0f1ec. [Reporting paper]() Recorded evaluation configuration: {"Technique":"Naïve Full-Context","Model":"LLaMA Maverick-17B-instruct"}. This record does not assert independent reproduction. ### Naïve Full-Context Answer Accuracy: 35. Result ID: 699688795698665ec4d0f1e8. [Reporting paper]() Recorded evaluation configuration: {"Technique":"Naïve Full-Context","Model":"GPT-4o"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.