Translation from the Information Bottleneck Perspective: an Efficiency Analysis of Spatial Prepositions in Bitexts
Not on the map. Why?
Abstract (arXiv)
Efficient communication requires balancing informativity and simplicity when encoding meanings. The Information Bottleneck (IB) framework captures this trade-off formally, predicting that natural language systems cluster near an optimal accuracy-complexity frontier. While supported in visual domains such as colour and motion, linguistic stimuli such as words in sentential context remain unexplored. We address this gap by framing translation as an IB optimisation problem, treating source sentences as stimuli and target sentences as compressed meanings. This allows IB analyses to be performed directly on bitexts rather than controlled naming experiments. We applied this to spatial prepositions across English, German and Serbian translations of a French novel. To estimate informativity, we conducted a pile-sorting pilot-study (N=35) and obtained similarity judgements of pairs of prepositions. We trained a low-rank projection model (D=5) that predicts these judgements (Spearman correlation: 0.78). Attested translations of prepositions lie closer to the IB optimal frontier than counterfactual alternatives, offering preliminary evidence that human translators exhibit communicative efficiency pressure in the spatial domain. More broadly, this work suggests that translation can serve as a window into the cognitive efficiency pressures shaping cross-linguistic semantic systems.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 12 of its 46 references that are in the map. First posted on arXiv 20 Mar 2026.
Builds on
12 of 46 references.
References · 12
- 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,907
- 2018WiC: the Word-in-Context Dataset for Evaluating Context-Sensitive Meaning Representationscited by 370
- 2014SimLex-999: Evaluating Semantic Models with (Genuine) Similarity Estimationcited by 170
- 2021All Bark and No Bite: Rogue Dimensions in Transformer Language Models Obscure Representational Qualitycited by 150
- 2016The deterministic information bottleneckcited by 42
- 2020Multi-SimLex: A Large-Scale Evaluation of Multilingual and Cross-Lingual Lexical Semantic Similaritycited by 38
- 2020Contextualized Word Embeddings Encode Aspects of Human-Like Word Sense Knowledgecited by 18
- 2022Words are all you need? Language as an approximation for human similarity judgmentscited by 18
- 2020Extracting low-dimensional psychological representations from convolutional neural networkscited by 16
- 2024Comparing the Pearson and Spearman Correlation Coefficients Across Distributions and Sample Sizes: A Tutorial Using Simulations and Empirical Datacited by 16
- 2019CoSimLex: A Resource for Evaluating Graded Word Similarity in Contextcited by 13
- 2024Simpson's Paradox and the Accuracy-Fluency Tradeoff in Translationcited by 8
Built on by
Not cited yet.