SITE applies soft gradient-based head selection to inject ICL-derived task embeddings, outperforming prior embedding adaptation and few-shot ICL across generation, reasoning, and NLU tasks on 12 LLMs from 4B to 70B parameters.
Word Translation Without Parallel Data
6 Pith papers cite this work, alongside 252 external citations. Polarity classification is still indexing.
6
Pith papers citing it
252
external citations · Pith
abstract
State-of-the-art methods for learning cross-lingual word embeddings have relied on bilingual dictionaries or parallel corpora. Recent studies showed that the need for parallel data supervision can be alleviated with character-level information. While these methods showed encouraging results, they are not on par with their supervised counterparts and are limited to pairs of languages sharing a common alphabet. In this work, we show that we can build a bilingual dictionary between two languages without using any parallel corpora, by aligning monolingual word embedding spaces in an unsupervised way. Without using any character information, our model even outperforms existing supervised methods on cross-lingual tasks for some language pairs. Our experiments demonstrate that our method works very well also for distant language pairs, like English-Russian or English-Chinese. We finally describe experiments on the English-Esperanto low-resource language pair, on which there only exists a limited amount of parallel data, to show the potential impact of our method in fully unsupervised machine translation. Our code, embeddings and dictionaries are publicly available.
representative citing papers
CodeBLEU improves correlation with human programmer scores on code synthesis tasks by adding syntactic AST matching and semantic data-flow matching to the standard BLEU n-gram approach.
Cross-lingual SSD recovers affective dimensions across languages with broad alignment but structured residual differences in arousal and dominance linked to bodily threat, authority, and other factors.
A framework extracts embeddings from Korean-English bilingual LLMs across thousands of spaces and uses similarity distributions plus logistic classifiers to identify lexical gaps with AUCs of 0.81 and 0.76.
Token-to-Token alignment rephrases prompts into shared structure then matches token embeddings by semantic similarity, making linear interpolation a meaningful operation for blending in text-to-image models.
Machine interpreting should shift from fidelity metrics to three design priorities—agency, grounding, and experience—drawn from interpreting studies to close the usability gap with human-mediated communication.
citing papers explorer
-
Soft Head Selection for Injecting ICL-Derived Task Embeddings
SITE applies soft gradient-based head selection to inject ICL-derived task embeddings, outperforming prior embedding adaptation and few-shot ICL across generation, reasoning, and NLU tasks on 12 LLMs from 4B to 70B parameters.
-
CodeBLEU: a Method for Automatic Evaluation of Code Synthesis
CodeBLEU improves correlation with human programmer scores on code synthesis tasks by adding syntactic AST matching and semantic data-flow matching to the standard BLEU n-gram approach.
-
Supervised Semantic Differential for Cross-Cultural Concept Analysis: A Case Study of Human Affect
Cross-lingual SSD recovers affective dimensions across languages with broad alignment but structured residual differences in arousal and dominance linked to bodily threat, authority, and other factors.
-
Discovering Lexical Gaps Using Embeddings from Multilingual LLMs
A framework extracts embeddings from Korean-English bilingual LLMs across thousands of spaces and uses similarity distributions plus logistic classifiers to identify lexical gaps with AUCs of 0.81 and 0.76.
-
Token-to-Token Alignment of Text Embeddings for Semantic Blending
Token-to-Token alignment rephrases prompts into shared structure then matches token embeddings by semantic similarity, making linear interpolation a meaningful operation for blending in text-to-image models.
-
Bridging the Usability Gap: Lessons from Interpreting Studies for Machine Interpreting Design
Machine interpreting should shift from fidelity metrics to three design priorities—agency, grounding, and experience—drawn from interpreting studies to close the usability gap with human-mediated communication.