# Implicit Sentiment Analysis on SemEval Restaurant ISA 2014

Source: [Implicit Sentiment Analysis on SemEval Restaurant ISA 2014](<https://www.sota2.com/research/sota/implicit-sentiment-analysis-on-semeval-restaurant-isa-2014>)

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

Data updated: 2026-02-26T05:42:45.701Z

Evaluation of Implicit Sentiment Analysis performance on the SemEval 2014 Restaurant dataset, measured by F1 score.

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 | Implicit Sentiment Analysis |
| Dataset (including any stated split) | SemEval Restaurant ISA 2014 |
| Metric | F1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 24 of 24 |
| 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

- [F1](<https://www.sota2.com/research/sota/implicit-sentiment-analysis-on-semeval-restaurant-isa-2014.md?metric=F1>): Higher is better; unit: PERCENT.

- [F1 (All)](<https://www.sota2.com/research/sota/implicit-sentiment-analysis-on-semeval-restaurant-isa-2014.md?metric=F1%20(All)>): Higher is better; unit: PERCENT.

- [F1 (ISA)](<https://www.sota2.com/research/sota/implicit-sentiment-analysis-on-semeval-restaurant-isa-2014.md?metric=F1%20(ISA)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 | Result date | Report paper |
| --- | --- | --- | --- |
| THOR | 79.73 | Not recorded | 68fa6fbc1e209131cea471d1 |
| THOR | 78.42 | Not recorded | 68fa6fbc1e209131cea471d1 |
| THOR | 77.68 | Not recorded | 68fa6fbc1e209131cea471d1 |
| Flan-T5+Prompt | 75.1 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERTAsp+SCAPT | 72.28 | Not recorded | 68fa6fbc1e209131cea471d1 |
| THOR | 71.7 | Not recorded | 68fa6fbc1e209131cea471d1 |
| Flan-T5+Prompt | 70.91 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERT+Prompt | 70.12 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERT+ISAIV | 69.66 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERT+RGAT | 67.79 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERTAsp+CEPT | 67.79 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERT+ADA | 65.92 | Not recorded | 68fa6fbc1e209131cea471d1 |
| BERT+SPC | 65.54 | Not recorded | 68fa6fbc1e209131cea471d1 |

### THOR

F1: 79.73. Result ID: 6990e8fa3a6c314849043761.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Implementation](<https://github.com/scofield7419/THOR-ISA>)

Recorded evaluation configuration: {"Model Parameters":"11B","Backbone":"Flan-T5","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### THOR

F1: 78.42. Result ID: 6990e8fa3a6c314849043763.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Implementation](<https://github.com/scofield7419/THOR-ISA>)

Recorded evaluation configuration: {"Model Parameters":"11B","Backbone":"Flan-T5","Evaluation Protocol":"Supervised fine-tuning","Method Variation":"w/o Reason-revising"}. This record does not assert independent reproduction.

### THOR

F1: 77.68. Result ID: 6990e8fa3a6c314849043762.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Implementation](<https://github.com/scofield7419/THOR-ISA>)

Recorded evaluation configuration: {"Model Parameters":"11B","Backbone":"Flan-T5","Evaluation Protocol":"Supervised fine-tuning","Method Variation":"w/o Self-consistency"}. This record does not assert independent reproduction.

### Flan-T5+Prompt

F1: 75.1. Result ID: 6990e8fa3a6c31484904375f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

Recorded evaluation configuration: {"Model Parameters":"11B","Backbone":"Flan-T5","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERTAsp+SCAPT

F1: 72.28. Result ID: 6990e8fa3a6c31484904375c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b291e209131cea2c6f1.md>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### THOR

F1: 71.7. Result ID: 6990e8fa3a6c314849043760.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Implementation](<https://github.com/scofield7419/THOR-ISA>)

Recorded evaluation configuration: {"Model Parameters":"250M","Backbone":"Flan-T5","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### Flan-T5+Prompt

F1: 70.91. Result ID: 6990e8fa3a6c31484904375e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

Recorded evaluation configuration: {"Model Parameters":"250M","Backbone":"Flan-T5","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERT+Prompt

F1: 70.12. Result ID: 6990e8fa3a6c31484904375d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERT+ISAIV

F1: 69.66. Result ID: 6990e8fa3a6c31484904375b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d391e209131cea388bf.md>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERT+RGAT

F1: 67.79. Result ID: 6990e8fa3a6c314849043759.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa676c1e209131cea160cd.md>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERTAsp+CEPT

F1: 67.79. Result ID: 6990e8fa3a6c31484904375a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b291e209131cea2c6f1.md>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERT+ADA

F1: 65.92. Result ID: 6990e8fa3a6c314849043758.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66ab1e209131cea11894.md>)

[Implementation](<https://github.com/deepopinion/domain-adapted-atsc>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

### BERT+SPC

F1: 65.54. Result ID: 6990e8fa3a6c314849043757.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6fbc1e209131cea471d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"Model Parameters":"110M","Backbone":"BERT","Evaluation Protocol":"Supervised fine-tuning"}. This record does not assert independent reproduction.

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