# Multimodal Question Answering on WebQA Text

Source: [Multimodal Question Answering on WebQA Text](<https://www.sota2.com/research/sota/multimodal-question-answering-on-webqa-text>)

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

Data updated: 2026-06-11T08:05:28.008Z

Evaluation of model performance on the WebQA Text dataset for Multimodal Question Answering, reporting F1, Accuracy, R@k, and Token Count metrics.

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 | Multimodal Question Answering |
| Dataset (including any stated split) | WebQA Text |
| Metric | F1 Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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 Score](<https://www.sota2.com/research/sota/multimodal-question-answering-on-webqa-text.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/multimodal-question-answering-on-webqa-text.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [R@k](<https://www.sota2.com/research/sota/multimodal-question-answering-on-webqa-text.md?metric=R%40k>): Higher is better; unit: PERCENT.

- [Token Count](<https://www.sota2.com/research/sota/multimodal-question-answering-on-webqa-text.md?metric=Token%20Count>): Lower is better; unit: SCALAR.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| Full Context | 54.9 | Not recorded | 6a28f36e90966ade919c18ff |
| Qwen3-VL-8B | 52.3 | Not recorded | 6a28f36e90966ade919c18ff |
| Nemo-Emb | 52.1 | Not recorded | 6a28f36e90966ade919c18ff |
| BM25 Retrieval | 49.5 | Not recorded | 6a28f36e90966ade919c18ff |
| Nemo-Emb | 48.9 | Not recorded | 6a28f36e90966ade919c18ff |
| Qwen3-VL-8B | 47.6 | Not recorded | 6a28f36e90966ade919c18ff |
| Dense Retrieval | 45.2 | Not recorded | 6a28f36e90966ade919c18ff |
| BM25 Retrieval | 43.6 | Not recorded | 6a28f36e90966ade919c18ff |
| Nemo-Emb | 41.5 | Not recorded | 6a28f36e90966ade919c18ff |
| Qwen3-VL-8B | 40.7 | Not recorded | 6a28f36e90966ade919c18ff |
| Dense Retrieval | 39 | Not recorded | 6a28f36e90966ade919c18ff |
| BM25 Retrieval | 38.6 | Not recorded | 6a28f36e90966ade919c18ff |
| Dense Retrieval | 36.2 | Not recorded | 6a28f36e90966ade919c18ff |
| Latent Memory | 31.9 | Not recorded | 6a28f36e90966ade919c18ff |
| Latent Memory | 31.6 | Not recorded | 6a28f36e90966ade919c18ff |
| Latent Memory | 31.3 | Not recorded | 6a28f36e90966ade919c18ff |

### Full Context

F1 Score: 54.9. Result ID: 6a2a6ae0fbc373592c2752d7.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa778f1e209131cea73e1b.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval Method":"Full-context reference"}. This record does not assert independent reproduction.

### Qwen3-VL-8B

F1 Score: 52.3. Result ID: 6a2a6ae0fbc373592c2752e0.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6653fdd5c3617441a2.md>)

[Implementation](<https://github.com/QwenLM/Qwen3-VL-Embedding>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"5"}. This record does not assert independent reproduction.

### Nemo-Emb

F1 Score: 52.1. Result ID: 6a2a6ae0fbc373592c2752e3.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79f11e209131cea80280.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"5"}. This record does not assert independent reproduction.

### BM25 Retrieval

F1 Score: 49.5. Result ID: 6a2a6ae0fbc373592c2752da.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"5"}. This record does not assert independent reproduction.

### Nemo-Emb

F1 Score: 48.9. Result ID: 6a2a6ae0fbc373592c2752e2.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79f11e209131cea80280.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"2"}. This record does not assert independent reproduction.

### Qwen3-VL-8B

F1 Score: 47.6. Result ID: 6a2a6ae0fbc373592c2752df.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6653fdd5c3617441a2.md>)

[Implementation](<https://github.com/QwenLM/Qwen3-VL-Embedding>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"2"}. This record does not assert independent reproduction.

### Dense Retrieval

F1 Score: 45.2. Result ID: 6a2a6ae0fbc373592c2752dd.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa686d1e209131cea1c190.md>)

[Implementation](<https://github.com/facebookresearch/DPR>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"5"}. This record does not assert independent reproduction.

### BM25 Retrieval

F1 Score: 43.6. Result ID: 6a2a6ae0fbc373592c2752d9.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"2"}. This record does not assert independent reproduction.

### Nemo-Emb

F1 Score: 41.5. Result ID: 6a2a6ae0fbc373592c2752e1.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79f11e209131cea80280.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"1"}. This record does not assert independent reproduction.

### Qwen3-VL-8B

F1 Score: 40.7. Result ID: 6a2a6ae0fbc373592c2752de.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69608b6653fdd5c3617441a2.md>)

[Implementation](<https://github.com/QwenLM/Qwen3-VL-Embedding>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"1"}. This record does not assert independent reproduction.

### Dense Retrieval

F1 Score: 39. Result ID: 6a2a6ae0fbc373592c2752dc.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa686d1e209131cea1c190.md>)

[Implementation](<https://github.com/facebookresearch/DPR>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"2"}. This record does not assert independent reproduction.

### BM25 Retrieval

F1 Score: 38.6. Result ID: 6a2a6ae0fbc373592c2752d8.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"1"}. This record does not assert independent reproduction.

### Dense Retrieval

F1 Score: 36.2. Result ID: 6a2a6ae0fbc373592c2752db.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa686d1e209131cea1c190.md>)

[Implementation](<https://github.com/facebookresearch/DPR>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"1"}. This record does not assert independent reproduction.

### Latent Memory

F1 Score: 31.9. Result ID: 6a2a6ae0fbc373592c2752e5.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Implementation](<https://github.com/zz1358m/Latent-Memory-Master>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"2"}. This record does not assert independent reproduction.

### Latent Memory

F1 Score: 31.6. Result ID: 6a2a6ae0fbc373592c2752e4.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Implementation](<https://github.com/zz1358m/Latent-Memory-Master>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"1"}. This record does not assert independent reproduction.

### Latent Memory

F1 Score: 31.3. Result ID: 6a2a6ae0fbc373592c2752e6.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a28f36e90966ade919c18ff.md>)

[Implementation](<https://github.com/zz1358m/Latent-Memory-Master>)

Recorded evaluation configuration: {"Generator":"Gemma-3-12B-Instruct","Retrieval count (k)":"5"}. This record does not assert independent reproduction.

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