# Thermoelectric Property Extraction on 50-Paper Corpus

Source: [Thermoelectric Property Extraction on 50-Paper Corpus](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-10T07:30:21.074Z

Evaluation of models performing information extraction to identify and quantify thermoelectric properties (ZT, Seebeck coefficient, Thermal Conductivity) from a corpus of 50 scientific papers, measured using Precision, Recall, and F1-scores for each property.

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 | Thermoelectric property extraction |
| Dataset (including any stated split) | Thermoelectric property extraction 50-paper corpus |
| Metric | ZT Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [ZT Precision](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=ZT%20Precision>): Higher is better; unit: PERCENT.

- [ZT Recall](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=ZT%20Recall>): Higher is better; unit: PERCENT.

- [ZT F1-score](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=ZT%20F1-score>): Higher is better; unit: PERCENT.

- [Seebeck (S) Precision](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Seebeck%20(S)%20Precision>): Higher is better; unit: PERCENT.

- [Seebeck (S) Recall](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Seebeck%20(S)%20Recall>): Higher is better; unit: PERCENT.

- [Seebeck (S) F1-score](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Seebeck%20(S)%20F1-score>): Higher is better; unit: PERCENT.

- [Thermal Conductivity (κ) Precision](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Thermal%20Conductivity%20(%CE%BA)%20Precision>): Higher is better; unit: PERCENT.

- [Thermal Conductivity (κ) Recall](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Thermal%20Conductivity%20(%CE%BA)%20Recall>): Higher is better; unit: PERCENT.

- [Thermal Conductivity (κ) F1-score](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Thermal%20Conductivity%20(%CE%BA)%20F1-score>): Higher is better; unit: PERCENT.

- [Overall (Micro) Precision](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Overall%20(Micro)%20Precision>): Higher is better; unit: PERCENT.

- [Overall (Micro) Recall](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Overall%20(Micro)%20Recall>): Higher is better; unit: PERCENT.

- [Overall (Micro) F1-score](<https://www.sota2.com/research/sota/thermoelectric-property-extraction-on-50-paper-corpus.md?metric=Overall%20(Micro)%20F1-score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ZT Precision | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini 1.5 Pro | 92.4 | Not recorded | 68fa4c1a1e209131ce94aa90 |
| GPT-4.1 | 90.9 | Not recorded | 68fa4c1a1e209131ce94aa90 |
| Gemini 2.0 Flash | 89.7 | Not recorded | 68fa4c1a1e209131ce94aa90 |
| GPT-4.1 Mini | 88.5 | Not recorded | 68fa4c1a1e209131ce94aa90 |

### Gemini 1.5 Pro

ZT Precision: 92.4. Result ID: 69afc8176dbb7075b957bd62.

[Reporting paper](<https://www.sota2.com/research/paper/68fa4c1a1e209131ce94aa90.md>)

Recorded evaluation configuration: {"Model":"Gemini 1.5 Pro"}. This record does not assert independent reproduction.

### GPT-4.1

ZT Precision: 90.9. Result ID: 69afc8176dbb7075b957bd65.

[Reporting paper](<https://www.sota2.com/research/paper/68fa4c1a1e209131ce94aa90.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1"}. This record does not assert independent reproduction.

### Gemini 2.0 Flash

ZT Precision: 89.7. Result ID: 69afc8176dbb7075b957bd63.

[Reporting paper](<https://www.sota2.com/research/paper/68fa4c1a1e209131ce94aa90.md>)

Recorded evaluation configuration: {"Model":"Gemini 2.0 Flash"}. This record does not assert independent reproduction.

### GPT-4.1 Mini

ZT Precision: 88.5. Result ID: 69afc8176dbb7075b957bd64.

[Reporting paper](<https://www.sota2.com/research/paper/68fa4c1a1e209131ce94aa90.md>)

Recorded evaluation configuration: {"Model":"GPT-4.1 Mini"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
