Pith. sign in

REVIEW 4 major objections 6 minor 41 references

Mergenetic: a Simple Evolutionary Model Merging Library

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Mergenetic is an open-source library that makes evolutionary model merging cheap enough for a single consumer GPU and flexible enough to compare many optimization strategies.

desk verdict A genuinely useful library for evolutionary model merging, with case studies that overstate confidence; send it out but push for a sharper abstract and proper sensitivity analysis. read the letter →

arxiv 2505.11427 v1 pith:64VEC4WJ submitted 2025-05-16 cs.LG cs.AIcs.NE

Adrian Robert Minut , Tommaso Mencattini , Andrea Santilli , Donato Crisostomi , Emanuele Rodolà This is my paper · ORCID
classification cs.LGcs.AIcs.NE
keywords modelmergingevolutionaryalgorithmsmulti-objectiveoptimizationlargelanguagemodelsfitnessestimationcross-lingualtransferconsumerGPUopen-sourcelibrary
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

Mergenetic is an open-source library that frames the weights of a model merge as variables in an evolutionary search: an optimizer proposes merge coefficients, the library builds the merged checkpoint, and a cheap fitness estimate scores it, repeating the loop over generations. The paper argues this is the first general-purpose library for evolutionary model merging in language models, letting users mix merging strategies and optimization algorithms freely instead of being locked into one routine. The central benefit is cost: fitness is computed on small subsamples or with statistical estimators rather than full benchmark runs, so meaningful searches fit on a single consumer GPU. Two demonstrations support the claim: merging four language-specific fine-tuned models into one multilingual model that beats every single-language endpoint on ARC-Challenge by up to 19% accuracy, and merging an English math model with a Japanese-tuned model so the result improves Japanese math word-problem accuracy by 10-20% over either parent. If true, the contribution is a much lower barrier to a technique that previously had no flexible, accessible implementation.

What carries the argument

The load-bearing component is the evolutionary search loop embodied in the searcher and problem classes: a genotype is a vector of merge coefficients, the merger layer converts that genotype into a checkpoint, and the evaluator converts the checkpoint into fitness scores that can be computed on a subsample or estimated with IRT-based approximation. This loop is what lets arbitrary evolutionary algorithms be plugged in without rewriting the merging code, and it is also where the cost savings happen, because fitness, the most expensive step, is deliberately approximated.

What would settle it

Rerun both evolutionary searches with fitness computed on the full benchmarks (full ARC-Challenge per language and the full Japanese math benchmark) and compare the winning merge coefficients with those found on the small subsets; if the winning recipes diverge substantially, or if the subset-chosen merges lose most of their reported edge on the full test sets, the small-sample fitness assumption is falsified.

Watch

Extended reading notes

Core claim

The paper introduces Mergenetic as the first library purpose-built for evolutionary model merging in language models, combining six merging strategies with nineteen evolutionary algorithms and supporting both single- and multi-objective search. The central design is a modular search loop: a merger maps a genotype (the coefficients of a merge recipe) to a checkpoint, an evaluator maps that checkpoint to one or more fitness scores, and a searcher orchestrates selection, crossover, and mutation over generations. To keep the loop affordable, fitness is computed on subsampled datasets or through item-response-theory-style estimators instead of full benchmark evaluation. The paper reports that this pipeline produces a multilingual ARC model surpassing each of its four language-specific parents by up to 19% accuracy, and a Japanese math model that beats its English and Japanese parents by 10-20% on the Japanese math benchmark, using modest GPUs and only small fitness subsets (100 examples in the math case, 20 per language in the multilingual case).

Load-bearing premise

The load-bearing premise is that fitness computed on very small benchmark subsets (20 translated ARC examples per language and 100 Japanese math examples) ranks candidate merges the same way full-dataset evaluation would, and the paper provides no test of that correlation.

Editorial extensions

If this is right

  • Users can compare single-objective and multi-objective merging on the same pipeline by changing only the optimizer and the objective definitions, so questions like 'does balancing two languages beat optimizing one?' become configuration changes.
  • A practitioner with one consumer GPU and a set of already fine-tuned checkpoints can run an evolutionary search instead of hand-tuning merge weights or training a new model from scratch.
  • Merging becomes a way to transplant skills across languages, as the Japanese math experiment shows: an English math capability can be transferred into a Japanese-tuned model without retraining.
  • Because fitness is approximated, iterative workflows become practical: run a short search, inspect the logged population, adjust the objectives or the evaluation subset, and relaunch.

Reading between the lines

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

  • The small fitness subsets (20 examples per language and 100 math examples) are the main uncontrolled variable; a direct test would rerun the same searches with full-dataset fitness and compare the winning merge recipes, which would reveal whether the cost savings come with a quality cost.
  • The modular genotype-to-model-to-fitness loop should carry over to other domains with mergeable checkpoints and cheap evaluators, such as vision-language or multimodal models, but the paper demonstrates only text language models.
  • The built-in multi-objective support could be used to optimize per-language worst-case accuracy rather than average accuracy, turning the merge search into an explicitly fairness-aware procedure; the paper does not explore this direction.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper introduces Mergenetic, an open-source Python library for evolutionary model merging built on MergeKit and PyMoo, with a Python API, CLI, and GUI. It supports six merging strategies, nineteen evolutionary algorithms, custom fitness functions via LM-Eval-Harness, and lightweight estimators including IRT-based ones. The empirical section presents two case studies: a multilingual ARC-Challenge merge across Italian, English, German, and Dutch using NSGA-II, and a cross-lingual math transfer to Japanese MGSM using a genetic algorithm. The abstract claims the library 'produces competitive results across tasks and languages using modest hardware.' The central claim is twofold: (i) the library architecture enables flexible, low-cost evolutionary merging, and (ii) the resulting merges are competitive with, and generally better than, their constituent models.

Significance. If the empirical claims were supported, Mergenetic would fill a real gap: no existing framework currently provides flexible, open-source support for composing merging strategies with evolutionary algorithms for LLMs on consumer GPUs. The modular design, permissive license, integration with established ecosystems (MergeKit, PyMoo, LM-Eval-Harness), and the inclusion of both CLI and GUI are concrete strengths, as is the fact that the demonstrations are reproducible from publicly available models and datasets. However, the evidence presented is not sufficient to establish the headline claim of 'competitive results': fitness is computed on very small subsamples, no baselines beyond constituent models are provided, no error bars or repeated seeds are reported, and the quantitative claims are not consistently supported by the figures. The paper would be stronger as a systems/software contribution with calibrated claims, or as an empirical paper with proper evaluation methodology.

major comments (4)
  1. [§5.1, Appendix A.2] The multilingual case study computes fitness on 20 translated ARC examples per language (Appendix A.2). With n=20, a single response changes accuracy by 5 percentage points, and the approximate 95% CI for an observed accuracy of 0.70 is about ±0.20, which is comparable to or larger than some reported gains (e.g., English: 0.79 vs 0.75). The NSGA-II run (population 25, 7 iterations) is not repeated, no full-ARC or held-out evaluation is reported, and no analysis shows that 20-example fitness ranks merges in the same order as full-dataset fitness. The claim in §5.1 that the merged model 'consistently outperforms' each constituent is therefore not established by the data as presented.
  2. [§5.2, Appendix A.1, Figure 5] The cross-lingual case study uses 100 MGSM-JA examples for fitness, in place of the 1,069 search points used by Akiba et al. (2025), but no sensitivity analysis or comparison with full-dataset fitness is provided. There are also no baselines such as equal-weight averaging, random search, or default TIES/DARE, so the claimed '10-20% accuracy improvement' in §5.2 cannot be attributed to evolutionary search rather than to the merging recipe itself. The numerical claim is also not clearly consistent with Figure 5: the visible deltas range from 0.18 (Abel, 0.24 to 0.42) to 0.41, which is neither 10 percentage points nor a uniform 10-20% relative improvement. The figure and text should be reconciled, and the claims re-stated with explicit error bars or repeated runs.
  3. [§5.1, §5.2, Appendix A.1–A.2] Neither case study reports repeated seeds, error bars, or significance tests, even though the search procedures are stochastic (initialization, SBX crossover, polynomial mutation). The final results are point estimates from a single evolutionary run, and it is unclear whether the reported model was selected based on dev-subset fitness or on final test-set performance. Given the small fitness subsets, the absence of variance estimates is a load-bearing gap for any quantitative claim about accuracy gains.
  4. [Abstract, §4.2.3, Appendix A.4] The abstract highlights 'lightweight fitness estimators to reduce evaluation costs,' but neither case study actually uses the IRT-based estimators (P-IRT, GP-IRT, MP-IRT, GMP-IRT) described in Appendix A.4; only plain random subsampling is used. No experiment demonstrates the accuracy or computational savings of these estimators within Mergenetic, nor compares them to full-dataset fitness. To substantiate the abstract claim, at least one case study should use an estimator and report correlation with full-dataset performance, or the claim should be narrowed to random subsampling.
minor comments (6)
  1. [Appendix A.3, Table 3] The SMS-EMOA row lists 'CTAEA' as its algorithm class; this appears to be a typo and should read 'SMS-EMOA'.
  2. [References] Several references are duplicated or incomplete: Deb et al. 2002a and 2002b are the same entry, Ilharco et al. appears as both 2022 and 2023, and Matena & Raffel and Stoica et al. lack publication years and venues.
  3. [Figure 5] The caption and bars should explicitly identify which bar is the merged model and which are the constituents; the label 'EvoLLM-JP-7B' is ambiguous because that name is also associated with the merged result.
  4. [§4.2.3] The term 'ConfigPE' is introduced without definition or citation; please spell it out or explain what it is.
  5. [§4.1, CLI example] The CLI example uses an en dash in '–eval-method' and '–merge-type'; these should be double hyphens ('--eval-method', '--merge-type').
  6. [Figure 1] The caption states that Mergenetic makes it easy to produce 'new state-of-the-art LLMs,' but the paper does not provide evidence for state-of-the-art performance; rephrase to avoid overclaiming.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's empirical claims are evaluated on held-out test sets, and the self-cited MERGE3 pipeline is used as a transparent reproduction rather than as a definitional input.

full rationale

Mergenetic is a systems/engineering contribution, so its 'derivation chain' is mostly a design narrative and a set of case studies rather than a mathematical derivation. The central empirical claims are supported by held-out evaluations: Section 4.2.4 states that test() evaluates the best solutions 'on an unseen test set,' and the two case studies in Section 5 and Appendix A.1/A.2 compute fitness on subsamples (100 MGSM-JA examples or 20 translated ARC examples) while reporting final performance on the respective test benchmarks. The reported accuracies are therefore not identical by construction to the fitness scores, so there is no fitted-input-called-prediction reduction. The paper does cite the authors' own prior work, notably MERGE3 (Mencattini et al., 2025), both as motivation and as one of the two pipelines reproduced, and Appendix A.4 credits the same predecessor for the MP-IRT/GMP-IRT estimators. However, the headline case studies use direct sub-sampled evaluation rather than those estimators, and the self-cited MERGE3 work is invoked as a published, externally evaluable pipeline rather than as an unverified uniqueness or existence theorem. Related-work self-citations (c2m3, ATM, Task Singular Vectors) are contextual and not load-bearing. The remaining concerns - e.g., the statistical adequacy of 20-100 example fitness subsets and the lack of sensitivity analysis - are empirical-validity issues, not circularity issues, and are appropriately handled under correctness risk rather than under this pass. Under the hard rule that circularity must be exhibited as a reduction or as a load-bearing self-citation chain, no such step is found in the manuscript.

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

The paper makes no theoretical derivation; its central claims rest on the validity of prior merging methods, on the reliability of small-subset and IRT-based fitness estimates, and on the correctness of the wrapped tools (MergeKit, PyMoo, LM-Eval-Harness). The experiments use hand-chosen subsample sizes and EA hyperparameters without sensitivity analysis.

free parameters (4)
  • fitness_subset_size_multilingual = 20 ARC examples per language
    Hand-chosen in Appendix A.2; no sensitivity analysis; the selected merge may change with a different subset.
  • fitness_subset_size_crosslingual = 100 MGSM-JA examples
    Hand-chosen in Appendix A.1, down from 1,069 in Akiba et al. (2025); no sensitivity analysis.
  • population_size = 25 (both case studies)
    Chosen in Appendices A.1 and A.2 without ablation.
  • generations = 7 (both case studies)
    Chosen in Appendices A.1 and A.2 without ablation.
assumptions (4)
  • domain assumption Fine-tuned models derived from a shared base model can be merged into a single model with combined capabilities using task arithmetic, TIES, DARE, or SLERP.
    The whole library is built on MergeKit's merging strategies; this is prior empirical knowledge, not proven in this paper.
  • domain assumption Performance on a small fitness subset (20 to 100 examples) is an adequate proxy for full-dataset performance during optimization.
    Used in both case studies; the paper gives no evidence or sensitivity analysis.
  • domain assumption IRT-based estimators (P-IRT, GP-IRT, MP-IRT, GMP-IRT) accurately approximate full-dataset correctness.
    Inherited from Mencattini et al. (2025) and Polo et al. (2024); Table 4 only gives qualitative ratings.
  • domain assumption PyMoo's evolutionary algorithms behave correctly when optimizing merging coefficients.
    The library delegates optimization to PyMoo without validating algorithm behavior on this search space.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mergenetic: a Simple Evolutionary Model Merging Library." pith.science (2026). https://pith.science/paper/64VEC4WJ

@misc{pith2026250511427,
  author       = {Pith},
  title        = {Pith review of: Mergenetic: a Simple Evolutionary Model Merging Library},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/64VEC4WJ}},
  note         = {Machine review of arXiv:2505.11427}
}
read the original abstract

Model merging allows combining the capabilities of existing models into a new one - post hoc, without additional training. This has made it increasingly popular thanks to its low cost and the availability of libraries that support merging on consumer GPUs. Recent work shows that pairing merging with evolutionary algorithms can boost performance, but no framework currently supports flexible experimentation with such strategies in language models. We introduce Mergenetic, an open-source library for evolutionary model merging. Mergenetic enables easy composition of merging methods and evolutionary algorithms while incorporating lightweight fitness estimators to reduce evaluation costs. We describe its design and demonstrate that Mergenetic produces competitive results across tasks and languages using modest hardware.

Figures

Figures reproduced from arXiv: 2505.11427 by the authors.

Figure 1
Figure 1. Mergenetic makes it easy to produce new state-of-the-art LLMs with minimal requirements. Recent research has shown that combining model merging with evolutionary algorithms can achieve superior performance (Akiba et al., 2025; Mencattini et al., 2025). However, this approach faces two key challenges: first, there is currently no library for experimenting with different evolu￾tionary algorithms and merging methods; s… view at source ↗
Figure 2
Figure 2. Example on how to use the Python API for power users who wish to customize merging workflows. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Screenshot of the Gradio-based GUI described in section [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Evolving a multi-lingual model spanning Ital [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 17 canonical work pages

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Abubakar Abid, Ali Abdalla, Ali Abid, Dawood Khan, Abdulrahman Alfozan, and James Zou. 2019. Gradio: Hassle-free sharing and testing of ml models in the wild. arXiv preprint arXiv:1906.02569

    arXiv 2019

  4. [4]

    Samuel Ainsworth, Jonathan Hayase, and Siddhartha Srinivasa. 2022. Git Re-Basin : Merging models modulo permutation symmetries. In The Eleventh International Conference on Learning Representations

    work page 2022

  5. [5]

    Takuya Akiba, Makoto Shing, Yujin Tang, Qi Sun, and David Ha. 2025. https://doi.org/10.1038/s42256-024-00975-8 Evolutionary optimization of model merging recipes . Nature Machine Intelligence

    doi:10.1038/s42256-024-00975-8 2025

  6. [6]

    Blank and K

    J. Blank and K. Deb . 2020. pymoo: Multi-objective optimization in python. IEEE Access, 8:89497--89509

    work page 2020

  7. [7]

    Thomas Bäck and Hans-Paul Schwefel. 1993. https://doi.org/10.1162/evco.1993.1.1.1 An overview of evolutionary algorithms for parameter optimization . Evolutionary Computation, 1(1):1--23

    doi:10.1162/evco.1993.1.1.1 1993

  8. [8]

    Leshem Choshen, Elad Venezian, Noam Slonim, and Yoav Katz. 2022. Fusing finetuned models for better pretraining. arXiv preprint arXiv:2204.03044

    arXiv 2022

Show all 41 references
  1. [9]

    Peter Clark, Isaac Cowhey, Oren Etzioni, Tushar Khot, Ashish Sabharwal, Carissa Schoenick, and Oyvind Tafjord. 2018. https://arxiv.org/abs/1803.05457 Think you have solved question answering? try arc, the AI2 reasoning challenge . CoRR, abs/1803.05457

    2018 arXiv

  2. [10]

    Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168

    2021 arXiv

  3. [11]

    Donato Crisostomi, Marco Fumero, Daniele Baieri, Florian Bernard, and Emanuele Rodolà. 2025. c^2m^3 : Cycle-consistent multi-model merging. In Advances in Neural Information Processing Systems, volume 37

    2025

  4. [12]

    Dipankar Dasgupta and Zbigniew Michalewicz. 1997. Evolutionary algorithms—an overview. Evolutionary algorithms in engineering applications, pages 3--28

    1997

  5. [13]

    MohammadReza Davari and Eugene Belilovsky. 2025. Model breadcrumbs: Scaling multi-task model merging with sparse masks. In European Conference on Computer Vision, pages 270--287. Springer

    2025

  6. [15]

    K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan. 2002 b . https://doi.org/10.1109/4235.996017 A fast and elitist multiobjective genetic algorithm: Nsga-ii . IEEE Transactions on Evolutionary Computation, 6(2):182--197

    2002

  7. [16]

    Kalyanmoy Deb, Karthik Sindhya, and Tatsuya Okabe. 2007. https://doi.org/10.1145/1276958.1277190 Self-adaptive simulated binary crossover for real-parameter optimization . In Proceedings of the 9th Annual Conference on Genetic and Evolutionary Computation, GECCO '07, page 1187...

    2007

  8. [17]

    Eiben and J.E

    A.E. Eiben and J.E. Smith. 2015. Introduction to Evolutionary Computing. Springer

    2015

  9. [18]

    Clémentine Fourrier, Nathan Habib, Alina Lozovskaya, Konrad Szafer, and Thomas Wolf. 2024. Open llm leaderboard v2. https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard

    2024

  10. [19]

    Leo Gao, Jonathan Tow, Baber Abbasi, Stella Biderman, Sid Black, Anthony DiPofi, Charles Foster, Laurence Golding, Jeffrey Hsu, Alain Le Noac'h, Haonan Li, Kyle McDonell, Niklas Muennighoff, Chris Ociepa, Jason Phang, Laria Reynolds, Hailey Schoelkopf, Aviya Skowron, Lintang S...

    2024 doi

  11. [20]

    Antonio Andrea Gargiulo, Donato Crisostomi, Maria Sofia Bucarelli, Simone Scardapane, Fabrizio Silvestri, and Emanuele Rodolà. 2025. https://arxiv.org/abs/2412.00081 Task singular vectors: Reducing task interference in model merging . Preprint, arXiv:2412.00081

    2025 arXiv

  12. [21]

    Charles Goddard, Shamane Siriwardhana, Malikeh Ehghaghi, Luke Meyers, Vladimir Karpukhin, Brian Benedict, Mark McQuade, and Jacob Solawetz. 2024. https://doi.org/10.18653/v1/2024.emnlp-industry.36 Arcee ' s M erge K it: A toolkit for merging large language models . In Proceedi...

    2024 doi

  13. [22]

    N. Hansen. 2023. https://arxiv.org/abs/1604.00772 The cma evolution strategy: A tutorial . arXiv preprint arXiv:1604.00772

    2023 arXiv

  14. [23]

    Ilharco, M.T

    G. Ilharco, M.T. Ribeiro, M. Wortsman, S. Gururangan, L. Schmidt, H. Hajishirzi, and A. Farhadi. 2023. https://arxiv.org/abs/2212.04089 Editing models with task arithmetic . arXiv preprint arXiv:2212.04089

    2023 arXiv

  15. [24]

    Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Suchin Gururangan, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi. 2022. Editing models with task arithmetic. The Eleventh International Conference on Learning Representations

    2022

  16. [25]

    Jiang, A

    A.Q. Jiang, A. Sablayrolles, A. Mensch, C. Bamford, D.S. Chaplot, D. de las Casas, F. Bressand, G. Lengyel, G. Lample, L. Saulnier, L.R. Lavaud, M.-A. Lachaux, P. Stock, T. Le Scao, T. Lavril, T. Wang, T. Lacroix, and W. El Sayed. 2023. https://arxiv.org/abs/2310.06825 Mistral...

    2023 arXiv

  17. [26]

    Joulin, E

    A. Joulin, E. Grave, P. Bojanowski, and T. Mikolov. 2016. Bag of tricks for efficient text classification. arXiv preprint arXiv:1607.01759

    2016 arXiv

  18. [27]

    Merging models with fisher-weighted averaging

    Michael Matena and Colin Raffel. Merging models with fisher-weighted averaging

  19. [28]

    Tommaso Mencattini, Adrian Robert Minut, Donato Crisostomi, Andrea Santilli, and Emanuele Rodolà. 2025. https://arxiv.org/abs/2502.10436 Merge ^3 : Efficient evolutionary merging on consumer-grade gpus . Preprint, arXiv:2502.10436

    2025 arXiv

  20. [29]

    Fidel A Guerrero Pe \ n a, Heitor Rapela Medeiros, Thomas Dubail, Masih Aminbeidokhti, Eric Granger, and Marco Pedersoli. 2023. Re-basin via implicit sinkhorn differentiation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20237--20246

    2023

  21. [30]

    Alain P \'e trowski and Sana Ben-Hamida. 2017. Evolutionary algorithms. John Wiley & Sons

    2017

  22. [31]

    Felipe Maia Polo, Lucas Weber, Leshem Choshen, Yuekai Sun, Gongjun Xu, and Mikhail Yurochkin. 2024. tinybenchmarks: evaluating llms with fewer examples. In Forty-first International Conference on Machine Learning

    2024

  23. [32]

    Esteban Real, Alok Aggarwal, Yanping Huang, and Quoc V Le. 2019. Regularized evolution for image classifier architecture search. In Proceedings of the aaai conference on artificial intelligence, volume 33, pages 4780--4789

    2019

  24. [33]

    Freda Shi, Mirac Suzgun, Markus Freitag, Xuezhi Wang, Suraj Srivats, Soroush Vosoughi, Hyung Won Chung, Yi Tay, Sebastian Ruder, Denny Zhou, et al. 2022. Language models are multilingual chain-of-thought reasoners. arXiv preprint arXiv:2210.03057

    2022 arXiv

  25. [34]

    Zipit! merging models from different tasks without training

    George Stoica, Daniel Bolya, Jakob Brandt Bjorner, Pratik Ramesh, Taylor Hearn, and Judy Hoffman. Zipit! merging models from different tasks without training. In The Twelfth International Conference on Learning Representations

  26. [35]

    Klaudia Thellmann, Bernhard Stadler, Michael Fromm, Jasper Schulze Buschhoff, Alex Jude, Fabio Barth, Johannes Leveling, Nicolas Flores-Herr, Joachim Köhler, René Jäkel, and Mehdi Ali. 2024. https://arxiv.org/abs/2410.08928 Towards cross-lingual llm evaluation for european lan...

    2024 arXiv

  27. [36]

    Amala Mary Vincent and P Jidesh. 2023. An improved hyperparameter optimization framework for automl systems using evolutionary algorithms. Scientific Reports, 13(1):4737

    2023

  28. [37]

    Ke Wang, Nikolaos Dimitriadis, Guillermo Ortiz-Jimenez, Fran c ois Fleuret, and Pascal Frossard. 2024. https://proceedings.mlr.press/v235/wang24k.html Localizing task information for improved model merging and compression . In Proceedings of the 41st International Conference o...

    2024

  29. [38]

    Mitchell Wortsman, Gabriel Ilharco, Samir Ya Gadre, Rebecca Roelofs, Raphael Gontijo-Lopes, Ari S Morcos, Hongseok Namkoong, Ali Farhadi, Yair Carmon, Simon Kornblith, and Ludwig Schmidt. 2022. https://proceedings.mlr.press/v162/wortsman22a.html Model soups: averaging weights ...

    2022

  30. [39]

    Prateek Yadav, Derek Tam, Leshem Choshen, Colin A Raffel, and Mohit Bansal. 2023. Ties-merging: Resolving interference when merging models. In Advances in Neural Information Processing Systems, volume 36, pages 7093--7115. Curran Associates, Inc

    2023

  31. [40]

    Enneng Yang, Li Shen, Guibing Guo, Xingwei Wang, Xiaochun Cao, Jie Zhang, and Dacheng Tao. 2024. Model merging in llms, mllms, and beyond: Methods, theories, applications and opportunities. arXiv preprint arXiv:2408.07666

    2024 arXiv

  32. [41]

    Le Yu, Bowen Yu, Haiyang Yu, Fei Huang, and Yongbin Li. 2024. https://proceedings.mlr.press/v235/yu24p.html Language models are super mario: Absorbing abilities from homologous models as a free lunch . In Proceedings of the 41st International Conference on Machine Learning, vo...

    2024

  33. [42]

    Luca Zhou, Daniele Solombrino, Donato Crisostomi, Maria Sofia Bucarelli, Fabrizio Silvestri, and Emanuele Rodol \`a . 2024. Atm: Improving model merging by alternating tuning and merging. arXiv preprint arXiv:2411.03055

    2024 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.