# Context Traceback on NarrativeQA LongBench

Source: [Context Traceback on NarrativeQA LongBench](<https://www.sota2.com/research/sota/context-traceback-on-narrativeqa-longbench>)

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

Data updated: 2026-04-14T07:05:04.455Z

Evaluation of model performance on the Context Traceback task using the NarrativeQA LongBench dataset, reporting Precision, Recall, and operational Cost.

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 | Context Traceback |
| Dataset (including any stated split) | NarrativeQA LongBench |
| Metric | Precision |
| Unit | PERCENT |
| 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

- [Precision](<https://www.sota2.com/research/sota/context-traceback-on-narrativeqa-longbench.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/context-traceback-on-narrativeqa-longbench.md?metric=Recall>): Higher is better; unit: PERCENT.

- [Cost (s)](<https://www.sota2.com/research/sota/context-traceback-on-narrativeqa-longbench.md?metric=Cost%20(s)>): Lower is better; unit: TIME.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| TracLLM | 96 | Not recorded | 68fa7a711e209131cea833bb |
| AttnTrace | 96 | Not recorded | 68fa7a711e209131cea833bb |
| STC | 95 | Not recorded | 68fa7a711e209131cea833bb |
| AT2 | 86 | Not recorded | 68fa7a711e209131cea833bb |
| LIME/CC | 78 | Not recorded | 68fa7a711e209131cea833bb |
| DAA | 72 | Not recorded | 68fa7a711e209131cea833bb |
| Shapley | 71 | Not recorded | 68fa7a711e209131cea833bb |
| Self-C | 25 | Not recorded | 68fa7a711e209131cea833bb |
| LOO | 21 | Not recorded | 68fa7a711e209131cea833bb |
| Gradient | 5 | Not recorded | 68fa7a711e209131cea833bb |

### TracLLM

Precision: 96. Result ID: 69dde70547f34ce924c4d604.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### AttnTrace

Precision: 96. Result ID: 69dde70547f34ce924c4d605.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### STC

Precision: 95. Result ID: 69dde70547f34ce924c4d600.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64dc1e209131cea06cb9.md>)

[Implementation](<https://github.com/pytorch/vision>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### AT2

Precision: 86. Result ID: 69dde70547f34ce924c4d5fe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78201e209131cea76e4b.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### LIME/CC

Precision: 78. Result ID: 69dde70547f34ce924c4d603.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74c91e209131cea63cc3.md>)

[Implementation](<https://github.com/MadryLab/context-cite>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### DAA

Precision: 72. Result ID: 69dde70547f34ce924c4d5fd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### Shapley

Precision: 71. Result ID: 69dde70547f34ce924c4d602.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63d01e209131cea00e17.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### Self-C

Precision: 25. Result ID: 69dde70547f34ce924c4d5ff.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cb71e209131cea35895.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### LOO

Precision: 21. Result ID: 69dde70547f34ce924c4d601.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

### Gradient

Precision: 5. Result ID: 69dde70547f34ce924c4d5fc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a711e209131cea833bb.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa614c1e209131ce9f38a9.md>)

Recorded evaluation configuration: {"LLM":"Llama-3.1-8B-Instruct"}. This record does not assert independent reproduction.

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