# Trajectory Prediction on ETH-UCY Top 1% hard tail samples

Source: [Trajectory Prediction on ETH-UCY Top 1% hard tail samples](<https://www.sota2.com/research/sota/trajectory-prediction-on-eth-ucy-top-1-percent-hard-tail-samples>)

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

Data updated: 2026-02-26T06:11:04.521Z

Evaluation of trajectory prediction models focusing specifically on the top 1% hardest samples (tail distribution) within the combined ETH and UCY datasets, measured using metrics like minADE and minFDE.

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 | Trajectory Prediction |
| Dataset (including any stated split) | ETH-UCY Top 1% hard tail samples |
| Metric | minADE |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 8 of 8 |
| 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

- [minADE](<https://www.sota2.com/research/sota/trajectory-prediction-on-eth-ucy-top-1-percent-hard-tail-samples.md?metric=minADE>): Lower is better; unit: SCALAR.

- [minFDE](<https://www.sota2.com/research/sota/trajectory-prediction-on-eth-ucy-top-1-percent-hard-tail-samples.md?metric=minFDE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | minADE | Result date | Report paper |
| --- | --- | --- | --- |
| T++E contrastive | 0.38 | Not recorded | 68fa6ee31e209131cea42a4d |
| T++E FEND | 0.38 | Not recorded | 68fa6ee31e209131cea42a4d |
| Traj++ EWTA+FEND | 0.84 | Not recorded | 68fa6ee31e209131cea42a4d |
| Traj++ EWTA+resample | 0.9 | Not recorded | 68fa6ee31e209131cea42a4d |
| Traj++ EWTA+LDAM | 0.92 | Not recorded | 68fa6ee31e209131cea42a4d |
| Traj++ EWTA+contrastive | 0.92 | Not recorded | 68fa6ee31e209131cea42a4d |
| Traj++ EWTA+reweighting | 0.97 | Not recorded | 68fa6ee31e209131cea42a4d |
| Traj++ EWTA | 0.98 | Not recorded | 68fa6ee31e209131cea42a4d |

### T++E contrastive

minADE: 0.38. Result ID: 699067c23a6c31484900500d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b961e209131cea2f128.md>)

Recorded evaluation configuration: {"base\_architecture":"Trajectron++","approach":"contrastive"}. This record does not assert independent reproduction.

### T++E FEND

minADE: 0.38. Result ID: 699067c23a6c31484900500e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Implementation](<https://github.com/ynw2021/FEND>)

Recorded evaluation configuration: {"base\_architecture":"Trajectron++","approach":"FEND"}. This record does not assert independent reproduction.

### Traj++ EWTA+FEND

minADE: 0.84. Result ID: 699067c23a6c31484900502f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Implementation](<https://github.com/ynw2021/FEND>)

Recorded evaluation configuration: {"Evaluation protocol":"best-of-20"}. This record does not assert independent reproduction.

### Traj++ EWTA+resample

minADE: 0.9. Result ID: 699067c23a6c31484900502b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa625f1e209131ce9f8df1.md>)

Recorded evaluation configuration: {"Evaluation protocol":"best-of-20"}. This record does not assert independent reproduction.

### Traj++ EWTA+LDAM

minADE: 0.92. Result ID: 699067c23a6c31484900502d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10e0f.md>)

[Implementation](<https://github.com/kaidic/LDAM-DRW>)

Recorded evaluation configuration: {"Evaluation protocol":"best-of-20"}. This record does not assert independent reproduction.

### Traj++ EWTA+contrastive

minADE: 0.92. Result ID: 699067c23a6c31484900502e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b961e209131cea2f128.md>)

Recorded evaluation configuration: {"Evaluation protocol":"best-of-20"}. This record does not assert independent reproduction.

### Traj++ EWTA+reweighting

minADE: 0.97. Result ID: 699067c23a6c31484900502c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654f1e209131cea0988a.md>)

[Implementation](<https://github.com/richardaecn/class-balanced-loss>)

Recorded evaluation configuration: {"Evaluation protocol":"best-of-20"}. This record does not assert independent reproduction.

### Traj++ EWTA

minADE: 0.98. Result ID: 699067c23a6c31484900502a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6ee31e209131cea42a4d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b961e209131cea2f128.md>)

Recorded evaluation configuration: {"Evaluation protocol":"best-of-20"}. This record does not assert independent reproduction.

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