Paper Citation Record · LEDGER

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging

As of 7 August 2026, this Paper Citation Record lists 82 of 82 outbound references and 2 inbound Pith citation observations for arXiv:2512.01461.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2512.01461 v2

Coverage vector

measured 82 of 82 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-03T19:16:04.999697Z

measured 84 of 84 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-07T06:34:17.273281+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-14T21:12:06.989077Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-05-14T21:12:58.892326Z

Reference resolution

82 of 82 outbound references displayed

verified exact0
verified fuzzy0
unresolved81
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation da399615-8fbb-4b24-b938-fe22c4bfd0ef · outbound

This paper cites Git re-basin: Merging models modulo permutation symmetries.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Git re-basin: Merging models modulo permutation symmetries

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:50.783729Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b07ffe1c-0c09-4324-a646-bb87d8b19463 · outbound

This paper cites Qwen Technical Report.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Qwen Technical Report

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:50.873340Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28dd684c-b709-4a27-964d-05dac9aaed0c · outbound

This paper cites Multitask learning.Machine learning, 28(1): 41–75, 1997.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Multitask learning.Machine learning, 28(1): 41–75, 1997

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:50.963658Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e992de7c-3f6b-450a-8b90-b0b43539e11a · outbound

This paper cites Semeval-2017 task 1: Semantic textual similarity multilingual and crosslingual focused evaluation.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Semeval-2017 task 1: Semantic textual similarity multilingual and crosslingual focused evaluation

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:51.095132Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation de5ec714-ec6f-4e0f-bc99-b65a1b40f4fc · outbound

This paper cites Fw-merging: Scaling model merging with frank-wolfe optimization.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Fw-merging: Scaling model merging with frank-wolfe optimization

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:51.252755Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 336b8f46-82db-4710-9d67-3687a94639ef · outbound

This paper cites Quora question pairs, 2018.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Quora question pairs, 2018

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:51.424928Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12efaed5-dd10-4d1d-ae86-4be31b08a428 · outbound

This paper cites Remote sensing image scene classification: Benchmark and state of the art.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Remote sensing image scene classification: Benchmark and state of the art

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:51.567201Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9fcd8eea-b3b3-4733-b9c9-78019ba3f713 · outbound

This paper cites Describing textures in the wild.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Describing textures in the wild

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:51.756694Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b11c9aa3-2bd8-431e-b30a-b9e113337182 · outbound

This paper cites Model breadcrumbs: scalable upcycling of finetuned foundation models via sparse task vectors merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Model breadcrumbs: scalable upcycling of finetuned foundation models via sparse task vectors merging

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:51.928142Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ddbf7887-15fd-4f7d-a1d2-ee1b2829481c · outbound

This paper cites The mnist database of handwritten digit images for machine learning research.IEEE signal processing magazine, pages 141–142, 2012.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging The mnist database of handwritten digit images for machine learning research.IEEE signal processing magazine, pages 141–142, 2012

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:52.105687Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6d345af-7335-4899-a85a-72d1b24c1874 · outbound

This paper cites Harmonizing and Merging Source Models for CLIP-based Domain Generalization.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Harmonizing and Merging Source Models for CLIP-based Domain Generalization

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:52.251953Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21fea34f-8013-43e0-a42d-cb2833372c9d · outbound

This paper cites Fine-Tuning Pretrained Language Models: Weight Initializations, Data Orders, and Early Stopping.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Fine-Tuning Pretrained Language Models: Weight Initializations, Data Orders, and Early Stopping

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:52.462406Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e046c146-1076-4729-a560-dee117416d40 · outbound

This paper cites Automatically constructing a corpus of sentential paraphrases.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Automatically constructing a corpus of sentential paraphrases

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:52.605455Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f29518c-dff4-4e75-94d7-1442fab13a4a · outbound

This paper cites Task singular vectors: Reducing task interference in model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Task singular vectors: Reducing task interference in model merging

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:52.747149Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 61197950-567a-430a-b54f-b8d35421d587 · outbound

This paper cites The third pascal recognizing textual entail- ment challenge.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging The third pascal recognizing textual entail- ment challenge

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:52.891962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a102c5cb-1c8b-4eda-a937-62edccec1e50 · outbound

This paper cites Gradient reweighting: Towards imbalanced class-incremental learning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Gradient reweighting: Towards imbalanced class-incremental learning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:53.067984Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf17e717-a55c-4d3b-8fa1-4f7368dc1168 · outbound

This paper cites an unresolved cited work.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Unresolved cited work

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:53.289663Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8f2e8975-5c84-4179-aad9-895c5c93b1a4 · outbound

This paper cites Measuring massive multitask language understanding.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Measuring massive multitask language understanding

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:53.468921Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b357f422-841c-42f5-8dd6-50e3593779be · outbound

This paper cites Emr-merging: Tuning-free high- performance model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Emr-merging: Tuning-free high- performance model merging

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:53.682709Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d3b1d99c-8781-4863-8a2e-0550319ebfef · outbound

This paper cites Multi-granular spatio-temporal token merging for training-free acceleration of video llms.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Multi-granular spatio-temporal token merging for training-free acceleration of video llms

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:53.885946Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c362a8e0-5616-4ef7-9792-5103f5a6c4bc · outbound

This paper cites Editing models with task arithmetic.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Editing models with task arithmetic

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:54.107569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d0af8a09-7bed-4837-922c-46280612bd72 · outbound

This paper cites Pytorch.Programming with TensorFlow: so- lution for edge computing applications, pages 87–104, 2021.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Pytorch.Programming with TensorFlow: so- lution for edge computing applications, pages 87–104, 2021

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:54.303434Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd2ce9cb-5cdf-4b88-9cf2-7eb6549eac0c · outbound

This paper cites Dataless knowledge fusion by merging weights of language models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Dataless knowledge fusion by merging weights of language models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:54.493367Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a456770-13d9-4b6f-bece-974294769e1e · outbound

This paper cites Repair: Renormalizing permuted activations for interpolation repair.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Repair: Renormalizing permuted activations for interpolation repair

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:54.780062Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df19c80f-66dc-40ad-abb3-a13669cfc5ee · outbound

This paper cites Task vector quantization for memory- efficient model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Task vector quantization for memory- efficient model merging

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:54.938422Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f3544bb3-a802-40fa-bb4c-c797cdc0f16e · outbound

This paper cites 3d object representations for fine-grained categorization.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging 3d object representations for fine-grained categorization

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.125850Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7052d903-a01a-474d-b5a8-9058ffcca397 · outbound

This paper cites Singular value decompo- sition.Numerical analysis for statisticians, pages 129–142,.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Singular value decompo- sition.Numerical analysis for statisticians, pages 129–142,

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.215038Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05e070d1-56fd-4993-a159-eabed10295ab · outbound

This paper cites Mitigating parameter interference in model merging via sharpness-aware fine-tuning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Mitigating parameter interference in model merging via sharpness-aware fine-tuning

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.302483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 034a1d56-e3f8-4bc8-bbcf-13b1e5d92356 · outbound

This paper cites Map: Low-compute model merging with amortized pareto fronts via quadratic approximation.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Map: Low-compute model merging with amortized pareto fronts via quadratic approximation

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.383126Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 52a0132b-2b31-4174-840b-1bfec81a6453 · outbound

This paper cites Model merging in pre-training of large language models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Model merging in pre-training of large language models

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.407891Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e4bced3-f827-4dee-9c27-6dcdab66b936 · outbound

This paper cites Truthfulqa: Measuring how models mimic human falsehoods.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Truthfulqa: Measuring how models mimic human falsehoods

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.473559Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 92ec3d1a-a429-409a-8b34-d11440dd8b04 · outbound

This paper cites 1bit-Merging: Dynamic Quantized Merging for Large Language Models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging 1bit-Merging: Dynamic Quantized Merging for Large Language Models

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.592279Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9c8dc007-c3d1-458d-bd2c-f43bb78300bf · outbound

This paper cites RoBERTa: A Robustly Optimized BERT Pretraining Approach.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging RoBERTa: A Robustly Optimized BERT Pretraining Approach

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.722729Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5af9a1b-a5b7-4cf3-8ce2-718930b74d08 · outbound

This paper cites Twin-merging: Dynamic integration of modular expertise in model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Twin-merging: Dynamic integration of modular expertise in model merging

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.843459Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a11c0d9d-1620-497e-bd79-55f369567d98 · outbound

This paper cites No task left behind: Isotropic model merging with common and task-specific subspaces.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging No task left behind: Isotropic model merging with common and task-specific subspaces

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:55.993993Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 498a2ceb-3d32-492d-ae8f-88f4761b69d2 · outbound

This paper cites Merging models with fisher-weighted averaging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Merging models with fisher-weighted averaging

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.131288Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c03961da-b987-40f1-bdb4-91d93cbc3706 · outbound

This paper cites Soft Merging of Experts with Adaptive Routing.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Soft Merging of Experts with Adaptive Routing

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.279870Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d8879641-fd29-461b-b695-df044e42c017 · outbound

This paper cites Reading digits in natural images with unsupervised feature learning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Reading digits in natural images with unsupervised feature learning

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.365931Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 729470e4-0a3d-43f8-b2ef-70fcd5d70d4a · outbound

This paper cites Dawin: Training-free dynamic weight interpolation for robust adaptation.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Dawin: Training-free dynamic weight interpolation for robust adaptation

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.476943Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a3a5bdce-b087-418c-b5f2-92d9a423c79b · outbound

This paper cites Accurate and efficient low-rank model merging in core space.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Accurate and efficient low-rank model merging in core space

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.604743Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3476cbb8-6728-4f84-8c35-0be6f294bf3c · outbound

This paper cites Bbq: A hand-built bias benchmark for question answering.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Bbq: A hand-built bias benchmark for question answering

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.685292Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb995de4-9539-408f-bfdf-8733495f60c9 · outbound

This paper cites Less is more: Efficient model merging with binary task switch.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Less is more: Efficient model merging with binary task switch

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.744947Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0fbedc3-a5b0-42a7-99ff-81e9cbcbe405 · outbound

This paper cites Mingle: Mixtures of null- space gated low-rank experts for test-time continual model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Mingle: Mixtures of null- space gated low-rank experts for test-time continual model merging

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.862391Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff65d78c-2bee-4ad8-889f-6c85b7262052 · outbound

This paper cites Language models are unsuper- vised multitask learners.OpenAI blog, page 9, 2019.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Language models are unsuper- vised multitask learners.OpenAI blog, page 9, 2019

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:56.948299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cae3866d-f1c6-4bc6-b408-f0faaadf5531 · outbound

This paper cites Learning transferable visual models from natural language supervision.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Learning transferable visual models from natural language supervision

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:57.071460Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1239da18-ec36-4d92-b49a-137665d5485f · outbound

This paper cites Squad: 100,000+ questions for machine comprehen- sion of text.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Squad: 100,000+ questions for machine comprehen- sion of text

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:57.194051Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ede5230-228f-4063-8972-68ede2ebd725 · outbound

This paper cites Recursive deep models for semantic compositionality over a sentiment treebank.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Recursive deep models for semantic compositionality over a sentiment treebank

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-03T19:15:57.670437Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66839cdb-60ed-45e8-a490-1e2813a7af8c · outbound

This paper cites The german traffic sign recognition benchmark: a multi-class classification competition.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging The german traffic sign recognition benchmark: a multi-class classification competition

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:00.424362Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6c41d9eb-5b8d-470f-a2a8-86f12c8bc214 · outbound

This paper cites Model merging with svd to tie the knots.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Model merging with svd to tie the knots

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:00.610131Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0d80eb1-baa0-43da-865e-fd4491aae670 · outbound

This paper cites Cat merging: A training-free approach for resolving conflicts in model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Cat merging: A training-free approach for resolving conflicts in model merging

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:00.740601Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 942444c8-e662-4279-88a3-4195dc3c13f3 · outbound

This paper cites Towards minimizing feature drift in model merging: Layer-wise task vector fusion for adaptive knowledge integration.arXiv preprint arXiv:2505.23859,.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Towards minimizing feature drift in model merging: Layer-wise task vector fusion for adaptive knowledge integration.arXiv preprint arXiv:2505.23859,

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:00.934327Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c71481e9-568d-47cb-991c-56ef88ffa65a · outbound

This paper cites Merging multi-task models via weight- ensembling mixture of experts.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Merging multi-task models via weight- ensembling mixture of experts

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:01.093238Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 34a8f182-8956-474d-b488-b29f84db6d57 · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:01.278500Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0cd60987-ede8-4b61-ab43-71b38f175ec0 · outbound

This paper cites Glue: A multi-task benchmark and analysis platform for natural language under- standing.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Glue: A multi-task benchmark and analysis platform for natural language under- standing

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:01.486664Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8150e4dd-4c6e-4f5f-869b-0df38ca90866 · outbound

This paper cites Localizing task infor- mation for improved model merging and compression.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Localizing task infor- mation for improved model merging and compression

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:01.696081Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51ee11c9-64d8-49d0-896a-aff217c76fd7 · outbound

This paper cites Neural network acceptability judgments.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Neural network acceptability judgments

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:01.794790Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c30a9d63-26a5-46d9-8e7b-75a06c9735f5 · outbound

This paper cites A broad-coverage challenge corpus for sentence understanding through inference.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging A broad-coverage challenge corpus for sentence understanding through inference

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:01.935010Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 174a7012-9e9d-42ff-bbec-9149ee20cd49 · outbound

This paper cites HuggingFace's Transformers: State-of-the-art Natural Language Processing.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging HuggingFace's Transformers: State-of-the-art Natural Language Processing

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:02.135379Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f240e85c-23a1-46eb-a660-daebd06cbf52 · outbound

This paper cites Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing in- ference time.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing in- ference time

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:02.337074Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c2d1c4e4-f8f6-4354-a062-694e60ca61ea · outbound

This paper cites Importance-based token merging for efficient image and video generation.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Importance-based token merging for efficient image and video generation

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:02.523080Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 213a6f0b-0eb5-4db6-80a5-c72eca9ced60 · outbound

This paper cites Sun database: Large-scale scene recog- nition from abbey to zoo.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Sun database: Large-scale scene recog- nition from abbey to zoo

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:02.757883Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 391f56a8-5815-4c73-a63d-cd86b227854e · outbound

This paper cites Ties-merging: Resolving interference when merging models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Ties-merging: Resolving interference when merging models

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:02.851393Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2b213b1d-01f6-4ef9-968e-152d905452f2 · outbound

This paper cites Calm: Consensus-aware localized merging for multi-task learning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Calm: Consensus-aware localized merging for multi-task learning

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.006217Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3622a0a5-e684-444d-b872-5a7c0238e7ce · outbound

This paper cites Adamerging: Adap- tive model merging for multi-task learning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Adamerging: Adap- tive model merging for multi-task learning

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.094983Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67bf4f89-086c-440e-81db-4a6f8af8f7f0 · outbound

This paper cites Continual model merg- ing without data: Dual projections for balancing stability and plasticity.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Continual model merg- ing without data: Dual projections for balancing stability and plasticity

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.188200Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4396b531-c3f7-4eeb-8c30-25a36dfc386c · outbound

This paper cites Mix data or merge models? balancing the helpfulness, honesty, and harmlessness of large language model via model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Mix data or merge models? balancing the helpfulness, honesty, and harmlessness of large language model via model merging

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.244174Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be94f38f-ed2a-4a70-b9de-248fdc155acf · outbound

This paper cites Merging Vision Transformers from Different Tasks and Domains.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Merging Vision Transformers from Different Tasks and Domains

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.288029Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 348c52c8-9e30-4b41-8a50-24c8f186c4d4 · outbound

This paper cites Language models are super mario: Absorbing abilities from homologous models as a free lunch.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Language models are super mario: Absorbing abilities from homologous models as a free lunch

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.380629Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7977c72d-a9ab-4e34-bd37-56107e8bc9b8 · outbound

This paper cites Robustmerge: Parameter-efficient model merging for mllms with direction robustness.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Robustmerge: Parameter-efficient model merging for mllms with direction robustness

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.483210Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9e70b426-2ff4-490b-99b5-ee44a0eeda04 · outbound

This paper cites An overview of multi-task learn- ing.National Science Review, pages 30–43, 2018.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging An overview of multi-task learn- ing.National Science Review, pages 30–43, 2018

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.635239Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1b00b4b1-9c16-4e57-bb62-7c766c93ea30 · outbound

This paper cites A survey on multi-task learning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging A survey on multi-task learning

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.727047Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2ac7eb7-36e6-482f-9c40-d6155dbbcff3 · outbound

This paper cites Beyond training: Dynamic token merging for zero-shot video understanding.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Beyond training: Dynamic token merging for zero-shot video understanding

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.810134Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2795d49-68cf-499e-a7f1-89948d08566e · outbound

This paper cites Merging loras like playing lego: Pushing the modularity of lora to extremes through rank-wise clustering.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Merging loras like playing lego: Pushing the modularity of lora to extremes through rank-wise clustering

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.870473Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 37368d30-7f3f-467c-b41c-a382c461aa48 · outbound

This paper cites Free-merging: Fourier transform for efficient model merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Free-merging: Fourier transform for efficient model merging

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:03.988962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5487cd89-048c-4cf7-b44c-597ac5832bb2 · outbound

This paper cites Aim: Adaptive inference of multi-modal llms via token merging and pruning.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Aim: Adaptive inference of multi-modal llms via token merging and pruning

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.061770Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2556686e-7b97-4b14-8979-60530e573964 · outbound

This paper cites Metagpt: Merging large language models using model exclusive task arithmetic.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Metagpt: Merging large language models using model exclusive task arithmetic

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.245221Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b8094bb2-fba1-41df-a3e1-caa9fa85152a · outbound

This paper cites Hm3: Hierarchical multi-objective model merging for pretrained models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Hm3: Hierarchical multi-objective model merging for pretrained models

Reference 77

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.347407Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d5aa789-7c83-4f32-9390-a487750bfb51 · outbound

This paper cites Remedy: Recipe merging dynam- ics in large vision-language models.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Remedy: Recipe merging dynam- ics in large vision-language models

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.534741Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b6b1bc46-69fc-4454-a575-24f6745b782f · outbound

This paper cites an unresolved cited work.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Unresolved cited work

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.715412Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cbd15223-86e5-4f2b-89b9-df7d0cda9260 · outbound

This paper cites Baselines for seen tasks • Individual Modelsrefer to task-specific models before merging.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Baselines for seen tasks • Individual Modelsrefer to task-specific models before merging

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.812594Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e61ff90-deff-4d4c-ae97-874850397e29 · outbound

This paper cites Datasets For visual classification tasks, we employ classification accu- racy as the evaluation metric.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging Datasets For visual classification tasks, we employ classification accu- racy as the evaluation metric

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-03T19:16:04.883595Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d3158880-0e22-4ba3-b934-41709a028a37 · outbound

This paper cites More Backbones In addition to the backbones evaluated in the main paper, we also assess the performance of various methods on ViT-B/16, ViT-L/14, and GPT-2 backbones.

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging More Backbones In addition to the backbones evaluated in the main paper, we also assess the performance of various methods on ViT-B/16, ViT-L/14, and GPT-2 backbones

Reference 82

Resolution: malformed identifier; no resolver link, observed 2026-08-03T19:16:04.999697Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation e522c0d3-3187-4312-9b12-9b693b87bc25 · inbound

Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion cites this paper.

Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-06-30T03:17:25.261573Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

Observation 19ecc26c-6536-4bfb-9a4c-3a81b4f2e090 · inbound

Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion cites this paper.

Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-06-30T03:17:25.261573Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.