MathNet delivers the largest multilingual Olympiad math dataset and benchmarks where models like Gemini-3.1-Pro reach 78% on solving but embedding models struggle on equivalent problem retrieval, with retrieval augmentation yielding up to 12% gains.
Language-agnostic BERT sentence embedding
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Hubness dominates as the causal driver of cross-lingual retrieval asymmetry in multilingual embeddings, with CSLS correction closing most of the reciprocity gap across five models and a 6518-item parallel corpus.
Embedding model performance on MTEB tasks correlates strongly with nearest-neighbor overlap and ICA magnitude differences in their embedding spaces.
A latent variable IRT framework decouples four safety-driving factors across 61 model configurations and 10 languages using 1.9 million evaluations, revealing that safety is largely unidimensional and that high cross-lingual gaps cluster in physical harm prompts and lower-resource languages.
Concept Separation Curves provide a classifier-independent method to visualize and quantify how sentence embeddings distinguish conceptual meaning from syntactic variations across languages and domains.
BLOOM is a 176B-parameter open-access multilingual language model trained on the ROOTS corpus that achieves competitive performance on benchmarks, with improved results after multitask prompted finetuning.
Machine translation preserves moral semantics in Polish social media data well enough for cross-lingual use, shown by 0.86 mean embedding similarity and 0.01-0.02 AUC gaps in moral foundations classification.
A new pre-training task that maps languages bidirectionally in embedding space improves machine translation by up to 11.9 BLEU, cross-lingual QA by 6.72 BERTScore points, and understanding accuracy by over 5% over strong baselines.
Larger 100K vocabularies in SPLADE models, especially those initialized with ESPLADE pretraining, improve retrieval effectiveness after pruning compared to 32K baselines while keeping similar efficiency.
citing papers explorer
-
Hubness, Not Anisotropy, Drives Cross-Lingual Retrieval Asymmetry in Multilingual Embedding Models
Hubness dominates as the causal driver of cross-lingual retrieval asymmetry in multilingual embeddings, with CSLS correction closing most of the reciprocity gap across five models and a 6518-item parallel corpus.
-
Structure Retention in Embedding Spaces as a Predictor of Benchmark Performance
Embedding model performance on MTEB tasks correlates strongly with nearest-neighbor overlap and ICA magnitude differences in their embedding spaces.
-
Finding Meaning in Embeddings: Concept Separation Curves
Concept Separation Curves provide a classifier-independent method to visualize and quantify how sentence embeddings distinguish conceptual meaning from syntactic variations across languages and domains.
-
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
BLOOM is a 176B-parameter open-access multilingual language model trained on the ROOTS corpus that achieves competitive performance on benchmarks, with improved results after multitask prompted finetuning.
-
Moral Semantics Survive Machine Translation: Cross-Lingual Evidence from Moral Foundations Corpora
Machine translation preserves moral semantics in Polish social media data well enough for cross-lingual use, shown by 0.86 mean embedding similarity and 0.01-0.02 AUC gaps in moral foundations classification.
-
Bridging Linguistic Gaps: Cross-Lingual Mapping in Pre-Training and Dataset for Enhanced Multilingual LLM Performance
A new pre-training task that maps languages bidirectionally in embedding space improves machine translation by up to 11.9 BLEU, cross-lingual QA by 6.72 BERTScore points, and understanding accuracy by over 5% over strong baselines.
-
The Role of Vocabularies in Learning Sparse Representations for Ranking
Larger 100K vocabularies in SPLADE models, especially those initialized with ESPLADE pretraining, improve retrieval effectiveness after pruning compared to 32K baselines while keeping similar efficiency.