Searching for Sound-Meaning Collisions: Graph-Based Affordance Retrieval and Multi-Evaluator Ranking for Pun Translation at CLEF 2026 JOKER Task 2

Adam Brikman; Prateek Awate; Russell Taylor

REVIEW 4 major objections 6 minor 20 references

Searching for Sound-Meaning Collisions: Graph-Based Affordance Retrieval and Multi-Evaluator Ranking for Pun Translation at CLEF 2026 JOKER Task 2

T0 review · 4 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Pun translation works by finding new sound-meaning collisions in the target language, not by preserving the original words.

desk verdict A competent system paper with a genuinely useful affordance-propagation analysis, but the headline claim waits on human evaluation the authors haven't run yet. read the letter →

arxiv 2608.04299 v1 pith:ED7ZCXAZ submitted 2026-08-05 cs.CL

Russell Taylor , Adam Brikman , Prateek Awate This is my paper · ORCID

classification cs.CL

keywords puntranslationcomputationalhumoraffordanceretrievalsound-meaningcollisionphoneticembeddingsretrieval-augmentedgenerationmulti-agentrankingwordplay

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper tries to show computationally that pun translation is a search for new places where the target language lets sound and meaning collide, rather than an attempt to preserve the source word. It builds a pipeline that retrieves candidate sound-meaning bridges, asks language models to generate multiple puns from those bridges, and ranks the candidates with several judging perspectives. Tracing how retrieved bridges survive each stage, the paper finds that generators actively use them, judges concentrate on the stronger ones, and exact sound matches win out of proportion to their availability. It also finds that retrieval is the main bottleneck: half of the source puns yield no usable bridge, and only a small fraction of final winners use retrieved material.

What carries the argument

The central object is the affordance: a retrieved pair of target-language items that connect the source pun's two semantic domains through a phonetic or semantic relation, giving a generator a place where new wordplay can be built. The argument is carried by a heterogeneous graph whose vertices are French words and expressions; semantic dense retrieval and learned phonetic embeddings define two kinds of edges, and bridge mining searches six ordered buckets for pairs that satisfy a phonetic similarity threshold. The same-sound and near-sound relations in the learned phonetic space, combined with a 355,803-entry phrase bank, let the system find phrase-level and approximate-sound bridges rather than only exact homophones. The machinery's work is to show that these bridges propagate through generation and selection, with their average quality rising as weaker affordances are pruned.

What would settle it

Collect human wordplay-quality ratings on a sample of the 4,061 translated puns, comparing the top-ranked ensemble with a plain single-model generation; if human judges do not prefer the affordance-guided ensemble, the claim that retrieval-driven sound-meaning discovery is responsible for pun translation success is not supported.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that successful pun translation emerges from a three-stage process of discovery, exploration, and selection, rather than from direct lexical transfer. The retrieval stage treats the French lexicon as a graph with semantic and phonetic edges and mines affordances, defined as pairs of target-language items that bridge the two semantic domains of the source pun through sound. Generators then expand a small set of retrieved affordances into multiple candidate puns, and a two-stage ranking architecture with comedian, linguist, editor, and translator personas compresses those candidates to a single winner. Across the pipeline, the affordances that survive to the final selection are fewer, stronger, and increasingly drawn from exact phonological collisions, even though such collisions are rare in the retrieval inventory. The paper reads this as evidence that the original proposal was correct, while conceding that the official leaderboard metric appears to reward preservation of source meaning more than humor and that human evaluation has not yet been run.

Load-bearing premise

The load-bearing assumption is that the official leaderboard score measures successful pun translation, but the paper itself notes the metric may reward preserving the source meaning over humor and that no human evaluation has yet been run.

Editorial extensions

If this is right

If retrieval-driven discovery is what makes pun translation work, then expanding coverage beyond the current 50.8 percent of source puns should directly raise the ceiling on final quality.
Because exact phonological collisions are selected at disproportionately high rates, retrieval systems should invest in finding more genuine homophones and near-homophones rather than only approximate rhymes.
Because judge personas disagree often and complete agreement is rare, rank aggregation across multiple quality perspectives is necessary to capture what makes a pun good.
The leaderboard's preference for translator-focused selection suggests that current automatic metrics may be measuring fidelity to the source joke rather than target-language wordplay, so final scores should not be read as humor quality.
The finding that a plain single-model generation beats the full pipeline on the leaderboard means generator capability is a limiting factor independent of retrieval and ranking.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural testable extension is to apply the same affordance-tracing analysis to languages with very different phonological systems; if the bottleneck persists there too, the difficulty may be structural rather than a matter of data scale.
The pipeline's emphasis on discovering collisions rather than preserving words could transfer to other creative-language tasks, such as metaphor or slogan generation, where the constraint is to re-found meaning in new sound-form material.
If the leaderboard rewards source-meaning preservation, then systems optimized on it may be learning the wrong objective for human humor; a human-evaluation study comparing the winning ensemble with a plain strong-model baseline would settle which story is right.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

The main thing you should know: this is a competent system paper with a genuinely useful empirical analysis, but its headline claim about successful pun translation is honestly conditional on human evaluation that hasn't happened yet.

What's new and good: the affordance propagation analysis. The authors trace 3,748 retrieved affordances through generation and selection (Tables 5–6) and show a clear trend: survivors are fewer and have higher mean quality, and exact homophones are overrepresented among winners (10.8% of the inventory versus 23.8–27.5% of winners). That's a concrete empirical pattern supporting Low's process model, and I haven't seen it shown this way before. The system extensions—phrase-level expression bank, learned phonetic embeddings, graph-based bridge mining, two-stage multi-judge ranking—are all sensible, and the ablations are useful.

The soft spots are real but not hidden. The authors themselves note in Section 4.7 that the automatic metric appears to reward source-meaning preservation over humor, and that the strongest single-model result came from a plain gpt-5.5 generation rather than the affordance pipeline. That undercuts the claim that the pipeline is driving leaderboard success. And since human evaluation hasn't been performed, the central story about sound–meaning affordances being selected because they're funnier is not yet externally supported. The stress-test concern lands: the affordance propagation could reflect optimization toward the metric rather than toward humanly funny puns. To the authors' credit, they say this explicitly in Section 4.7.

Minor issues: no code or data artifacts, no significance tests on the Table 5–6 proportions, and the leaderboard standings are preliminary with only their own runs visible. These are minor-to-moderate; they don't break the descriptive analysis, but they limit how strongly you can conclude.

Who it's for: people working on computational humor, pun translation, or retrieval-augmented generation for creative tasks. It's a useful case study and a reasonable extension of their 2025 system. It deserves a serious referee. The camera-ready version needs human evaluation results and ideally released artifacts. Without those, the main finding stays conditional.

I'd bring it to a reading group, but I wouldn't cite it as evidence yet.

Recommendation: accept for peer review, expect major revision.

Referee Report

4 major / 6 minor

Summary. The paper presents a system for English-to-French pun translation submitted to CLEF 2026 JOKER Task 2. It models translation as discovery, exploration, and selection: a retrieval stage searches semantic and phonetic neighborhoods of French expressions to identify "affordances" (sound-meaning bridges), LLM generators produce candidate puns conditioned on those affordances, and a multi-persona, two-stage Borda-ranking ensemble selects winners. The paper's primary contribution is an analysis of how retrieved affordances propagate through the pipeline, reporting that retained affordances become fewer in number but higher in mean quality, and that exact homophones are overrepresented among final winners. It concludes that successful pun translation emerges not from preserving source words but from discovering new places where sound and meaning collide.

Significance. If the central empirical claim were fully supported, this would be a useful computational realization of Low's theory of pun translation, and the per-stage retention statistics are a genuinely informative descriptive trace of a complex pipeline. The paper's strengths include the large-scale resource construction (a 370k-entry French expression bank and 4.46 million phonetic relations), the explicit six-stage bridge-mining decomposition, and the detailed retention tables, which are internally consistent. However, the external measure is a public leaderboard that the authors themselves say rewards source-meaning preservation more than humor, no human evaluation is reported, the method for detecting affordance usage in generated text is not specified, and no code, data, error bars, or significance tests are provided. The theoretical conclusion therefore currently exceeds the evidence supplied by the paper.

major comments (4)

[§3.5 and §4.7] The central claim—that successful pun translation emerges from discovering sound-meaning collisions—is evaluated only by the CLEF leaderboard (37.783), yet the authors state in §4.7 that "the official evaluation metric therefore appears to reward preservation of source meaning more strongly than creative pun generation" and in §3.5 that "human evaluation ... has not yet been performed." Because the metric may not track humor or wordplay quality, the affordance-selection results in Tables 5–6 cannot support the paper's main conclusion. Either human evaluation results should be added (as the authors promise for the camera-ready version) or the conclusion should be explicitly restricted to optimization of the current automatic metric.
[§3.2, Table 5] Table 5 shows that retrieved affordances appear in only 141 of 4,061 final winners (3.5%), and only 2,064 source puns (50.8%) yield affordances at all. Even among source puns with affordances, the winner-usage rate is only about 141/2,064 ≈ 6.8%. These numbers are difficult to reconcile with the abstract's claim that "successful pun translation emerges ... from discovering new places in the target language where sound and meaning collide." The paper should report a conditional analysis (e.g., winner quality with versus without affordance usage) or a no-retrieval baseline; otherwise the aggregate data suggest that affordances are optional for most successful outputs.
[§3.2, Tables 5 and 6] The entire propagation analysis rests on the detection of "affordance usage" in generated candidates, but the manuscript never defines how usage is determined. Is it exact surface matching, lemmatized matching, IPA matching, or semantic similarity to a retrieved affordance? Without a precise detection procedure, the retention percentages in Table 5 and the mean-score increases in Table 6 cannot be independently verified. Please specify the matching procedure and, ideally, provide the detection code or an error analysis.
[§3.2, Table 6] The claim that evaluators "progressively concentrate around stronger sound-meaning bridges" is based on small mean-score differences across stages (e.g., overall score 0.677 at retrieval versus 0.724 at winners; naturalness 0.520 versus 0.588). No confidence intervals, paired significance tests, or effect sizes are reported, so the reader cannot tell whether these differences are reliable or within sampling noise. Given the importance of this trend to the paper's thesis, it needs at least a paired test (for example, on puns with affordances present at every stage) and ideally bootstrap intervals.

minor comments (6)

[§1 author footnote] The corresponding author email contains an apparent text corruption ("envel⌢pe-⌢penrdtaylorjr@gatech.edu"); please correct it.
[§4.7] The statement that "the strongest single-model result was obtained by a single gpt-5.5 generation" should clarify whether the gpt-5.5 single-candidate run received retrieved affordances; §2.4 only describes affordance conditioning for claude-sonnet-4.6 and gemini-3-flash, so the text is ambiguous about what "plain" means here.
[§3.2] The sentence reporting "22.9% and 19.4% after Stage 1 selection, and 23.8% and 27.5% among Stage 2 finalists and winners" is ambiguous: specify that the first pair refers to Claude and Gemini and the second pair to finalists and winners, or restructure the sentence.
[§2.3.3 and Table 6] The "Overall" quality score in Table 6 is used as a central measure, but Section 2.3.3 does not give the formula or relative weights that combine the phonetic, semantic, naturalness, and pivot scores. Please provide the exact scoring function.
[References] Reference [5] is incomplete ("Wiktionary contributors, latex, 2026") and reference [11] appears garbled ("Xiv–2109. Ar"); both should be corrected. Also, the paper does not mention a data/code availability link, which would substantially aid reproducibility of the retention analysis.
[§5] The final sentence of the conclusion is missing a period ("... lies in finding them").

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central affordance claims rest on external leaderboard evaluation and independent held-out retrieval checks; self-citations are background, not load-bearing.

full rationale

The load-bearing results are anchored to an external benchmark (Section 3.5 reports a public-leaderboard score of 37.783) and to held-out intrinsic evaluations (Section 2.2.2's phonetic-retrieval benchmark with 6,321 held-out queries), rather than to parameters fitted to the claimed outcome. The affordance-propagation statistics in Tables 5-6 are descriptive traces of a fixed pipeline: retrieval outputs are matched post hoc against generated candidates and finalists, and the reported increases in mean affordance quality are observations about that pipeline, not equations that define the conclusion in terms of its inputs. The paper itself flags the key validity caveat in Section 3.5 ('Human evaluation ... has not yet been performed') and Section 4.7 ('current automatic metrics underweight the qualities that make target-language puns entertaining'), which is an external-validity risk, not circularity. The authors cite their own 2025 system in Section 1, but only as background; it does not carry the argument. No fitted parameter is renamed as a prediction, and no uniqueness theorem is imported from prior work by the same authors. The only mild concern is that 'generators actively exploit retrieved opportunities' is a manipulation check on a prompt that supplies those affordances, but the paper does not use that observation alone to derive its central conclusion.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central conclusions rest on four domain assumptions: Low's framework, the affordance-usage attribution, LLM-judge validity, and leaderboard validity. The last is explicitly flagged by the authors. No formal axioms are needed beyond these domain assumptions.

free parameters (5)

Phonetic similarity threshold tau
In the bridge mining definition, pairs with cosine(ipa embeddings) >= tau count as phonetic bridges; tau determines the affordance inventory but its value is not reported.
Per-surface and per-root caps
Used to prevent one lexical family from dominating affordance lists; exact cap values are not given.
Persona and judge weights in Borda aggregation = equal, 45/30/25/10, and persona-only variants in Table 8
Weights are chosen by hand and materially change winner distributions (Table 8).
Borda top-k and point weights
Rankings are aggregated with weighted Borda voting over top-five lists, but the specific point schedule is not stated.
Expression mining thresholds
OpenSubtitles collocations used frequency thresholds and PMI scoring with spaCy; exact cutoffs are not reported.

assumptions (4)

domain assumption Low's theoretical framework, that pun translation is best modeled as searching for new sound-meaning collisions rather than preserving words, is the correct basis for a computational system.
The paper operationalizes this theory without testing it against alternative translation strategies; Sections 1 and 2 cite Low [1] as foundational.
domain assumption Generated candidates can be reliably detected as using a specific retrieved affordance via the matching procedure.
All propagation statistics in Section 3.2 depend on this attribution, but the matching algorithm and thresholds are not described.
domain assumption LLM judge personas produce rankings that track humor, wordplay, fluency, and intent.
Section 2.5 builds ranking on persona prompts and Borda aggregation without human calibration; the paper itself notes evaluators disagree and may be position-biased.
domain assumption The public leaderboard score is a meaningful proxy for pun translation quality.
Section 3.5 and 4.7 state human evaluation has not been performed and the automatic metric may reward meaning preservation over humor, so this assumption is explicitly in doubt.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Searching for Sound-Meaning Collisions: Graph-Based Affordance Retrieval and Multi-Evaluator Ranking for Pun Translation at CLEF 2026 JOKER Task 2." pith.science (2026). https://pith.science/paper/ED7ZCXAZ

@misc{pith2026260804299,
  author       = {Pith},
  title        = {Pith review of: Searching for Sound-Meaning Collisions: Graph-Based Affordance Retrieval and Multi-Evaluator Ranking for Pun Translation at CLEF 2026 JOKER Task 2},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ED7ZCXAZ}},
  note         = {Machine review of arXiv:2608.04299}
}

read the original abstract

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.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 11 canonical work pages

[1]

P. A. Low, Translating jokes and puns, Perspectives: Studies in Translatology 19 (2011) 59–70

work page 2011
[2]

Taylor, B

R. Taylor, B. Herbert, M. Sana, Pun intended: Multi-agent translation of wordplay with contrastive learning and phonetic-semantic embeddings for clef joker 2025 task 2, in: Working Notes of CLEF 2025 – Conference and Labs of the Evaluation Forum, volume 4038 ofCEUR Workshop Proceedings, CEUR-WS.org, 2025. URL: https://ceur-ws.org/Vol-4038/paper_229.pdf

work page 2025
[3]

J. J. Gibson, The Ecological Approach to Visual Perception, Houghton Mifflin, Boston, 1979

work page 1979
[4]

T. Miller, The punster’s amanuensis: The proper place of humans and machines in the translation of wordplay, in: Proceedings of the Human-Informed Translation and Interpreting Technology Workshop (HiT-IT), Incoma, 2019, 2019, pp. 57–65

work page 2019
[5]

URL: https://en.wiktionary.org/wiki/latex

Wiktionary contributors, latex, 2026. URL: https://en.wiktionary.org/wiki/latex

work page 2026
[6]

Savary, C

A. Savary, C. Ramisch, S. Cordeiro, F. Sangati, V. Vincze, B. QasemiZadeh, M. Candito, F. Cap, V. Giouli, I. Stoyanova, A. Doucet, The parseme shared task on automatic identification of verbal multiword expressions, in: Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017), Association for Computational Linguistics, 2017, pp. 31–47. URL: ht...

doi:10.18653/v1/w17-1704 2017
[7]

Lison, J

P. Lison, J. Tiedemann, M. Kouylekov, Opensubtitles2018: Statistical rescoring of sentence align- ments in large, noisy parallel corpora, in: Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), European Language Resources Association (ELRA), Miyazaki, Japan, 2018

work page 2018
[8]

Honnibal, I

M. Honnibal, I. Montani, S. Van Landeghem, A. Boyd, spacy: Industrial-strength natural language processing in python, 2020. URL: https://doi.org/10.5281/zenodo.1212303. doi:10.5281/zenodo. 1212303

doi:10.5281/zenodo.1212303 2020

Show all 20 references

[9]

J. Chen, S. Xiao, P. Zhang, K. Luo, D. Lian, Z. Liu, M3-embedding: Multi-linguality, multi- functionality, multi-granularity text embeddings through self-knowledge distillation, in: L.-W. Ku, A. Martins, V. Srikumar (Eds.), Findings of the Association for Computational Linguis...

2024 doi
[10]

Johnson, M

J. Johnson, M. Douze, H. Jégou, Billion-scale similarity search with gpus, arXiv preprint arXiv:1702.08734 (2017)

2017 arXiv
[11]

Sharma, et al., Phonetic word embeddings, arXiv e-prints (2021) Xiv–2109

R. Sharma, et al., Phonetic word embeddings, arXiv e-prints (2021) Xiv–2109. Ar

2021
[12]

Reimers, I

N. Reimers, I. Gurevych, Making monolingual sentence embeddings multilingual using knowledge distillation, in: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, Association for Computational Linguistics, 2020. URL: https://arxiv.org/abs/2004.09813

2020 arXiv
[13]

Grave, P

E. Grave, P. Bojanowski, P. Gupta, A. Joulin, T. Mikolov, Learning word vectors for 157 languages, in: Proceedings of the International Conference on Language Resources and Evaluation (LREC 2018), 2018

2018
[14]

Wang, et al., Mmte: Corpus and metrics for evaluating machine translation quality of metaphor- ical language, 2024.arXiv:2406.13698, arXiv preprint

S. Wang, et al., Mmte: Corpus and metrics for evaluating machine translation quality of metaphor- ical language, 2024.arXiv:2406.13698, arXiv preprint

2024 arXiv
[15]

Góes, et al., Crowd score: A method for the evaluation of jokes using large language model ai voters as judges, 2022.arXiv:2212.11214, arXiv preprint

F. Góes, et al., Crowd score: A method for the evaluation of jokes using large language model ai voters as judges, 2022.arXiv:2212.11214, arXiv preprint

2022 arXiv
[16]

D. G. Saari, Basic geometry of voting, Springer (1995)

1995
[17]

M. G. Kendall, A new measure of rank correlation, Biometrika 30 (1938) 81–93. doi: 10.1093/ biomet/30.1-2.81

1938
[18]

S. Zhong, et al., Let’s think outside the box: Exploring leap-of-thought in large language models with creative humor generation, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

2024
[19]

Wang, et al., Innovative thinking, infinite humor: Humor research of large language models through structured thought leaps, 2024.arXiv:2410.10370, arXiv preprint

H. Wang, et al., Innovative thinking, infinite humor: Humor research of large language models through structured thought leaps, 2024.arXiv:2410.10370, arXiv preprint

2024 arXiv
[20]

URL: http://www

PACE, Partnership for an Advanced Computing Environment (PACE), 2017. URL: http://www. pace.gatech.edu

2017

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

P. A. Low, Translating jokes and puns, Perspectives: Studies in Translatology 19 (2011) 59–70

work page 2011

[2] [2]

Taylor, B

R. Taylor, B. Herbert, M. Sana, Pun intended: Multi-agent translation of wordplay with contrastive learning and phonetic-semantic embeddings for clef joker 2025 task 2, in: Working Notes of CLEF 2025 – Conference and Labs of the Evaluation Forum, volume 4038 ofCEUR Workshop Proceedings, CEUR-WS.org, 2025. URL: https://ceur-ws.org/Vol-4038/paper_229.pdf

work page 2025

[3] [3]

J. J. Gibson, The Ecological Approach to Visual Perception, Houghton Mifflin, Boston, 1979

work page 1979

[4] [4]

T. Miller, The punster’s amanuensis: The proper place of humans and machines in the translation of wordplay, in: Proceedings of the Human-Informed Translation and Interpreting Technology Workshop (HiT-IT), Incoma, 2019, 2019, pp. 57–65

work page 2019

[5] [5]

URL: https://en.wiktionary.org/wiki/latex

Wiktionary contributors, latex, 2026. URL: https://en.wiktionary.org/wiki/latex

work page 2026

[6] [6]

Savary, C

A. Savary, C. Ramisch, S. Cordeiro, F. Sangati, V. Vincze, B. QasemiZadeh, M. Candito, F. Cap, V. Giouli, I. Stoyanova, A. Doucet, The parseme shared task on automatic identification of verbal multiword expressions, in: Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017), Association for Computational Linguistics, 2017, pp. 31–47. URL: ht...

doi:10.18653/v1/w17-1704 2017

[7] [7]

Lison, J

P. Lison, J. Tiedemann, M. Kouylekov, Opensubtitles2018: Statistical rescoring of sentence align- ments in large, noisy parallel corpora, in: Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), European Language Resources Association (ELRA), Miyazaki, Japan, 2018

work page 2018

[8] [8]

Honnibal, I

M. Honnibal, I. Montani, S. Van Landeghem, A. Boyd, spacy: Industrial-strength natural language processing in python, 2020. URL: https://doi.org/10.5281/zenodo.1212303. doi:10.5281/zenodo. 1212303

doi:10.5281/zenodo.1212303 2020