# Backward Tracing on OVO-Bench Reactive QA 1.0 (test)

Source: [Backward Tracing on OVO-Bench Reactive QA 1.0 (test)](<https://www.sota2.com/research/sota/backward-tracing-on-ovo-bench-reactive-qa-1-0-test>)

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

Data updated: 2026-02-26T05:41:49.871Z

Evaluation of model performance on the Backward Tracing task using the test split of OVO-Bench Reactive QA version 1.0, measured by EPM, ASI, and HLD.

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 | Backward Tracing |
| Dataset (including any stated split) | OVO-Bench Reactive QA 1.0 (test) |
| Metric | EPM |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [EPM](<https://www.sota2.com/research/sota/backward-tracing-on-ovo-bench-reactive-qa-1-0-test.md?metric=EPM>): Higher is better; unit: SCALAR.

- [ASI](<https://www.sota2.com/research/sota/backward-tracing-on-ovo-bench-reactive-qa-1-0-test.md?metric=ASI>): Higher is better; unit: SCALAR.

- [HLD](<https://www.sota2.com/research/sota/backward-tracing-on-ovo-bench-reactive-qa-1-0-test.md?metric=HLD>): Lower is better; unit: SCALAR.

## Recorded results

| Method | EPM | Result date | Report paper |
| --- | --- | --- | --- |
| ROMA | 56.57 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| ROMA | 56.57 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| ROMA | 55.89 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| ROMA | 55.22 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| ROMA | 54.88 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| ROMA | 53.87 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| Dispider | 48.48 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| Flash-VStream-7B | 39.06 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| VideoLLM-online | 22.22 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |
| MMDuet | 10.44 | Not recorded | 6969c5efe25ad2fa9d1ec7f7 |

### ROMA

EPM: 56.57. Result ID: 6997b92e5698665ec40fa879.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

Recorded evaluation configuration: {"Aggregation layers (K)":1}. This record does not assert independent reproduction.

### ROMA

EPM: 56.57. Result ID: 6997b92e5698665ec40fa87b.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

Recorded evaluation configuration: {"Positive loss weight (wpos)":4}. This record does not assert independent reproduction.

### ROMA

EPM: 55.89. Result ID: 6997b92e5698665ec40fa876.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ROMA

EPM: 55.22. Result ID: 6997b92e5698665ec40fa877.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

Recorded evaluation configuration: {"Training strategy":"Mixed Training"}. This record does not assert independent reproduction.

### ROMA

EPM: 54.88. Result ID: 6997b92e5698665ec40fa87a.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

Recorded evaluation configuration: {"Positive loss weight (wpos)":2}. This record does not assert independent reproduction.

### ROMA

EPM: 53.87. Result ID: 6997b92e5698665ec40fa878.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

Recorded evaluation configuration: {"Architecture configuration (Speak Head)":"Without Speak Head"}. This record does not assert independent reproduction.

### Dispider

EPM: 48.48. Result ID: 6997b92e5698665ec40fa874.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767f1e209131cea6d744.md>)

[Implementation](<https://github.com/Mark12Ding/Dispider>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Flash-VStream-7B

EPM: 39.06. Result ID: 6997b92e5698665ec40fa875.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73ea1e209131cea5ede6.md>)

[Implementation](<https://invinciblwyq.github.io/vstream-page>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### VideoLLM-online

EPM: 22.22. Result ID: 6997b92e5698665ec40fa872.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73a71e209131cea5db14.md>)

[Implementation](<https://www.sota2.com/showlab.github.io/videollm-online>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MMDuet

EPM: 10.44. Result ID: 6997b92e5698665ec40fa873.

[Reporting paper](<https://www.sota2.com/research/paper/6969c5efe25ad2fa9d1ec7f7.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75e81e209131cea6a921.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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