Searching for Sound-Meaning Collisions: Graph-Based Affordance Retrieval and Multi-Evaluator Ranking for Pun Translation at CLEF 2026 JOKER Task 2
Quick Answer
This paper explores computational pun translation by modeling it as a discovery process, emphasizing sound-meaning collisions over word equivalence.
Quick Take
A retrieval system identifies phonological and semantic affordances, while multiple language models generate and rank translations. Findings indicate that successful translations arise from discovering new sound-meaning connections rather than preserving source words, highlighting retrieval as a key challenge.
Key Points
- Models sound-meaning collisions as key to effective pun translation.
- Retrieval system searches for phonological and semantic affordances.
- Multiple language models generate competing translations for evaluation.
- Findings suggest retrieval bottlenecks hinder computational pun translation.
- Successful translations emerge from discovering new target language connections.
DeepSignal Analysis
What happened
The paper investigates computational pun translation by modeling it as a discovery process focused on sound-meaning connections. A retrieval system identifies potential phonological and semantic affordances, while multiple language models generate and rank translations. The findings suggest that successful translations depend on discovering new sound-meaning relationships rather than merely preserving source words.
Key evidence
- The study emphasizes that pun translators should seek new sound-meaning connections rather than equivalent words, as proposed by Low fifteen years ago.
- A retrieval system is used to search for target-language affordances, which are defined as sound-meaning bridges that support new wordplay.
- The analysis reveals that many puns do not yield usable affordances, indicating that retrieval is a significant challenge in computational pun translation.
Why it matters
This research highlights a shift in approach for pun translation, moving away from direct word equivalence to a focus on creative sound-meaning interactions. By identifying and utilizing these connections, the study aims to enhance the effectiveness of computational translation systems. Understanding the retrieval process is crucial, as it remains a bottleneck in achieving successful pun translations.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Fifteen years ago, Low proposed that pun translators should stop searching for equivalent words and instead search for new points of contact between sound and meaning. In this paper, we investigate that idea computationally. We model pun translation as a process of discovery, exploration, and selection. A retrieval system searches semantic and phonological neighborhoods for target-language affordances: sound-meaning bridges that may support new wordplay. Multiple language models then explore these opportunities by generating competing translations, while a multi-perspective generate-and-rank architecture selects among them. Beyond system development, our primary contribution is an analysis of how retrieved affordances propagate through the translation process. We find that generators actively exploit retrieved opportunities, evaluators progressively concentrate around stronger sound-meaning bridges, and exact phonological collisions are selected at disproportionately high rates when available. At the same time, many puns still yield no usable affordances, suggesting that retrieval remains the central bottleneck in computational pun translation. The resulting picture is remarkably close to the process envisioned by Low. Successful pun translation emerges not from preserving source-language words, but from discovering new places in the target language where sound and meaning collide.
| Comments: | CLEF 2026 Working Notes, 21-24 September 2026, Jena, Germany |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2608.04299 [cs.CL] |
| (or arXiv:2608.04299v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2608.04299 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Russell Taylor [view email]
[v1]
Wed, 5 Aug 2026 00:06:56 UTC (40 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.