bkai-foundation-models/vietnamese-bi-encoder

Explore bkai-foundation-models/vietnamese-bi-encoder: model specifications, capabilities, benchmark results, platform pricing and original sources.

huggingface/bkai-foundation-models/vietnamese-bi-encoder

Specifications & capabilities

Published model ID
vietnamese-bi-encoder

Evaluations

1–6 / 6

MTEB

69.966%

source id
mteb
leaderboard name
MTEB Semantic Similarity
dimension
embedding
metric
Average task score
rank
335
sample size
1
aggregation scope
splits
  • test
subsets
  • default
task_count
1
task_names
  • STSBenchmark-VN
sample_size
1
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 84f9d9ada0d1a3c37557398b9ae9fcedcdf40be0
source model id
bkai-foundation-models/vietnamese-bi-encoder
Original result ↗

MTEB

58.595%

source id
mteb
leaderboard name
MTEB Classification
dimension
embedding
metric
Average task score
rank
350
sample size
14
aggregation scope
splits
  • test
  • validation
subsets
  • default
task_count
13
task_names
  • AmazonCounterfactualVNClassification
  • AmazonPolarityVNClassification
  • AmazonReviewsVNClassification
  • Banking77VNClassification
  • EmotionVNClassification
  • ImdbVNClassification
  • MTOPDomainVNClassification
  • MTOPIntentVNClassification
  • MassiveIntentVNClassification
  • MassiveScenarioVNClassification
  • ToxicConversationsVNClassification
  • TweetSentimentExtractionVNClassification
  • VieStudentFeedbackClassification
sample_size
14
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 84f9d9ada0d1a3c37557398b9ae9fcedcdf40be0
source model id
bkai-foundation-models/vietnamese-bi-encoder
Original result ↗

MTEB

34.127%

source id
mteb
leaderboard name
MTEB Clustering
dimension
embedding
metric
Average task score
rank
496
sample size
5
aggregation scope
splits
  • test
subsets
  • default
task_count
5
task_names
  • RedditClustering-VN
  • RedditClusteringP2P-VN
  • StackExchangeClustering-VN
  • StackExchangeClusteringP2P-VN
  • TwentyNewsgroupsClustering-VN
sample_size
5
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 84f9d9ada0d1a3c37557398b9ae9fcedcdf40be0
source model id
bkai-foundation-models/vietnamese-bi-encoder
Original result ↗

MTEB

35.809%

source id
mteb
leaderboard name
MTEB Other
dimension
embedding
metric
Average task score
rank
707
sample size
33
aggregation scope
splits
  • dev
  • test
  • validation
subsets
  • default
task_count
32
task_names
  • ArguAna-VN
  • AskUbuntuDupQuestions-VN
  • BIOSSES-VN
  • CQADupstackAndroid-VN
  • CQADupstackGis-VN
  • CQADupstackMathematica-VN
  • CQADupstackPhysics-VN
  • CQADupstackProgrammers-VN
  • CQADupstackStats-VN
  • CQADupstackTex-VN
  • CQADupstackUnix-VN
  • CQADupstackWebmasters-VN
  • CQADupstackWordpress-VN
  • ClimateFEVER-VN
  • DBPedia-VN
  • FEVER-VN
  • FiQA2018-VN
  • HotpotQA-VN
  • MSMARCO-VN
  • NFCorpus-VN
  • NQ-VN
  • Quora-VN
  • SCIDOCS-VN
  • SICK-R-VN
  • SciDocsRR-VN
  • SciFact-VN
  • SprintDuplicateQuestions-VN
  • StackOverflowDupQuestions-VN
  • TRECCOVID-VN
  • Touche2020-VN
  • TwitterSemEval2015-VN
  • TwitterURLCorpus-VN
sample_size
33
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 84f9d9ada0d1a3c37557398b9ae9fcedcdf40be0
source model id
bkai-foundation-models/vietnamese-bi-encoder
Original result ↗

MTEB

29.546%

source id
mteb
leaderboard name
MTEB Retrieval
dimension
embedding
metric
Average task score
rank
730
sample size
2
aggregation scope
splits
  • test
  • validation
subsets
  • default
task_count
2
task_names
  • GreenNodeTableMarkdownRetrieval
  • VieQuADRetrieval
sample_size
2
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 84f9d9ada0d1a3c37557398b9ae9fcedcdf40be0
source model id
bkai-foundation-models/vietnamese-bi-encoder
Original result ↗

MTEB

41.647%

source id
mteb
leaderboard name
MTEB Overall
dimension
embedding
metric
Average task score
rank
737
sample size
55
aggregation scope
splits
  • dev
  • test
  • validation
subsets
  • default
task_count
53
task_names
  • AmazonCounterfactualVNClassification
  • AmazonPolarityVNClassification
  • AmazonReviewsVNClassification
  • ArguAna-VN
  • AskUbuntuDupQuestions-VN
  • BIOSSES-VN
  • Banking77VNClassification
  • CQADupstackAndroid-VN
  • CQADupstackGis-VN
  • CQADupstackMathematica-VN
  • CQADupstackPhysics-VN
  • CQADupstackProgrammers-VN
  • CQADupstackStats-VN
  • CQADupstackTex-VN
  • CQADupstackUnix-VN
  • CQADupstackWebmasters-VN
  • CQADupstackWordpress-VN
  • ClimateFEVER-VN
  • DBPedia-VN
  • EmotionVNClassification
  • FEVER-VN
  • FiQA2018-VN
  • GreenNodeTableMarkdownRetrieval
  • HotpotQA-VN
  • ImdbVNClassification
  • MSMARCO-VN
  • MTOPDomainVNClassification
  • MTOPIntentVNClassification
  • MassiveIntentVNClassification
  • MassiveScenarioVNClassification
  • NFCorpus-VN
  • NQ-VN
  • Quora-VN
  • RedditClustering-VN
  • RedditClusteringP2P-VN
  • SCIDOCS-VN
  • SICK-R-VN
  • STSBenchmark-VN
  • SciDocsRR-VN
  • SciFact-VN
  • SprintDuplicateQuestions-VN
  • StackExchangeClustering-VN
  • StackExchangeClusteringP2P-VN
  • StackOverflowDupQuestions-VN
  • TRECCOVID-VN
  • Touche2020-VN
  • ToxicConversationsVNClassification
  • TweetSentimentExtractionVNClassification
  • TwentyNewsgroupsClustering-VN
  • TwitterSemEval2015-VN
  • TwitterURLCorpus-VN
  • VieQuADRetrieval
  • VieStudentFeedbackClassification
sample_size
55
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 84f9d9ada0d1a3c37557398b9ae9fcedcdf40be0
source model id
bkai-foundation-models/vietnamese-bi-encoder
Original result ↗

Original sources

1–4 / 4

Hugging Face · bkai-foundation-models/vietnamese-bi-encoder ↗

arXiv · Paper · 2403.01616 ↗

Hugging Face · Catalog ↗

Hugging Face · Catalog ↗

Source links

1–2 / 2

Hugging Face model card

1–1 / 1

Explore more models

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace