# LKS

Source: [LKS](<https://www.sota2.com/research/dataset/lks>)

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

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.

## Benchmarks

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

- [Model Merging on TED Talks and XLSum 40 sequential tasks (test)](<https://www.sota2.com/research/sota/model-merging-on-ted-talks-and-xlsum-40-sequential-tasks-test.md>): Model Merging; TED Talks and XLSum 40 sequential tasks (test).

- [Likes Prediction on TED Talks](<https://www.sota2.com/research/sota/likes-prediction-on-ted-talks.md>): Likes Prediction; TED Talks.

- [Regression predicting Views on TED talks dataset](<https://www.sota2.com/research/sota/regression-predicting-views-on-ted-talks-dataset.md>): Regression predicting Views; TED talks dataset.

- [Multiple Object Forecasting on Citywalks (3 train-test splits)](<https://www.sota2.com/research/sota/multiple-object-forecasting-on-citywalks-3-train-test-splits.md>): Multiple Object Forecasting; Citywalks (3 train-test splits).

- [Synthetic Long-context Reasoning on GraphWalks](<https://www.sota2.com/research/sota/synthetic-long-context-reasoning-on-graphwalks.md>): Synthetic Long-context Reasoning; GraphWalks.

- [WSI Classification on LKS](<https://www.sota2.com/research/sota/wsi-classification-on-lks.md>): WSI Classification; LKS.

- [Long Context Performance on Graphwalks Bfs (\<128K)](<https://www.sota2.com/research/sota/long-context-performance-on-graphwalks-bfs-128k.md>): Long Context Performance; Graphwalks Bfs (\<128K).

- [Drag-style image editing on Ted-talks (test)](<https://www.sota2.com/research/sota/drag-style-image-editing-on-ted-talks-test.md>): Drag-style image editing; Ted-talks (test).

- [Graph traversal on GraphWalks](<https://www.sota2.com/research/sota/graph-traversal-on-graphwalks.md>): Graph traversal; GraphWalks.

- [Human Image Animation on TED-talks (test)](<https://www.sota2.com/research/sota/human-image-animation-on-ted-talks-test.md>): Human Image Animation; TED-talks (test).

- [Robot Navigation on Real-robot urban sidewalks closed-loop (test)](<https://www.sota2.com/research/sota/robot-navigation-on-real-robot-urban-sidewalks-closed-loop-test.md>): Robot Navigation; Real-robot urban sidewalks closed-loop (test).

- [Video Reenactment on TED-talks dataset (test)](<https://www.sota2.com/research/sota/video-reenactment-on-ted-talks-dataset-test.md>): Video Reenactment; TED-talks dataset (test).

- [Long-context graph reasoning on GraphWalks (F1)](<https://www.sota2.com/research/sota/long-context-graph-reasoning-on-graphwalks-f1.md>): Long-context graph reasoning; GraphWalks.

- [2D Character Animation on TED-talks dataset](<https://www.sota2.com/research/sota/2d-character-animation-on-ted-talks-dataset.md>): 2D Character Animation; TED-talks dataset.

- [Long Context Performance on Graphwalks Parents](<https://www.sota2.com/research/sota/long-context-performance-on-graphwalks-parents.md>): Long Context Performance; Graphwalks Parents.

- [Long-context Language Modeling on Composite Suite (MRCR v2, GraphWalks, LongBench v2, RULER, AA-LCR)](<https://www.sota2.com/research/sota/long-context-language-modeling-on-composite-suite-mrcr-v2-graphwalks-longbench-v2-ruler-aa-lcr.md>): Long-context language modeling; Long-Context Evaluation Suite MRCR v2, GraphWalks, LongBench v2, RULER, AA-LCR.

- [Long Context Performance on Graphwalks Parents (\<128K)](<https://www.sota2.com/research/sota/long-context-performance-on-graphwalks-parents-128k.md>): Long Context Performance; Graphwalks Parents (\<128K).

- [Code-switching mitigation on TED Talks English-Hindi Llama-3.2-1B n=500 (test)](<https://www.sota2.com/research/sota/code-switching-mitigation-on-ted-talks-english-hindi-llama-3-2-1b-n-500-test.md>): Code-switching mitigation; TED Talks English-Hindi Llama-3.2-1B n=500 (test).

- [Code-switching mitigation on TED Talks English-Russian Llama-3.2-1B (test)](<https://www.sota2.com/research/sota/code-switching-mitigation-on-ted-talks-english-russian-llama-3-2-1b-test.md>): Code-switching mitigation; TED Talks English-Russian Llama-3.2-1B, n=500 (test).

- [Code-switching mitigation on TED Talks English-Spanish Llama-3.2-1B, n=500 (test)](<https://www.sota2.com/research/sota/code-switching-mitigation-on-ted-talks-english-spanish-llama-3-2-1b-n-500-test.md>): Code-switching mitigation; TED Talks English-Spanish Llama-3.2-1B, n=500 (test).

- [Code-switching mitigation on TED Talks English-Chinese Llama-3.2-1B (test)](<https://www.sota2.com/research/sota/code-switching-mitigation-on-ted-talks-english-chinese-llama-3-2-1b-test.md>): Code-switching mitigation; TED Talks English-Chinese Llama-3.2-1B, n=500 (test).

- [Machine Translation on TED Talks Arabic-English ar-en (test)](<https://www.sota2.com/research/sota/machine-translation-on-ted-talks-arabic-english-ar-en-test.md>): Machine Translation; TED Talks Arabic-English ar-en (test).

- [Machine Translation on TED Talks Hebrew-English (he-en) (test)](<https://www.sota2.com/research/sota/machine-translation-on-ted-talks-hebrew-english-he-en-test.md>): Machine Translation; TED Talks Hebrew-English (he-en) (test).

- [Machine Translation on TED Talks Italian-English (it-en) (test)](<https://www.sota2.com/research/sota/machine-translation-on-ted-talks-italian-english-it-en-test.md>): Machine Translation; TED Talks Italian-English it-en (test).

- [Machine Translation on TED Talks German-English (de-en) (test)](<https://www.sota2.com/research/sota/machine-translation-on-ted-talks-german-english-de-en-test.md>): Machine Translation; TED Talks German-English (de-en) (test).

- [Speech-driven 3D Holistic Body Motion Generation on TalkSHOW Perceptual Study Set 1.0 (test)](<https://www.sota2.com/research/sota/speech-driven-3d-holistic-body-motion-generation-on-talkshow-perceptual-study-set-1-0-test.md>): Speech-driven 3D Holistic Body Motion Generation; TalkSHOW Perceptual Study Set 1.0 (test).

- [Video reconstruction on TED-talks](<https://www.sota2.com/research/sota/video-reconstruction-on-ted-talks.md>): Video reconstruction; TED-talks.

- [Video reconstruction on TED-talks (test)](<https://www.sota2.com/research/sota/video-reconstruction-on-ted-talks-test.md>): Video reconstruction; TED-talks (test).

- [Automatic Speech Recognition on TED-talks 2012 (test)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ted-talks-2012-test.md>): Automatic Speech Recognition; TED-talks 2012 (test).

- [Automatic Speech Recognition on TED-talks 2011 (TST)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-ted-talks-2011-tst.md>): Automatic Speech Recognition; TED-talks 2011 (TST).

- [Image Animation on TED-talks (test)](<https://www.sota2.com/research/sota/image-animation-on-ted-talks-test.md>): Image Animation; TED-talks (test).

- [Animation Quality on TED-talks (user study)](<https://www.sota2.com/research/sota/animation-quality-on-ted-talks-user-study.md>): Animation Quality; TED-talks (user study).