Paper Citation Record · LEDGER

MARS: Unleashing the Power of Variance Reduction for Training Large Models

As of 23 August 2026, this Paper Citation Record lists 95 of 95 outbound references and 20 inbound Pith citation observations for arXiv:2411.10438.

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

pith.paper-citation-record.v1
2411.10438 v4

Coverage vector

measured 95 of 95 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-12T19:43:02.115828Z

measured 115 of 115 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-22T06:32:14.747728+00:00

measured 20 of 20 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-10T17:20:51.021889Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-03T20:18:56.218978Z

Reference resolution

95 of 95 outbound references displayed

verified exact2
verified fuzzy23
unresolved70
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 32587236-7371-4bc6-8cae-c4c9317b1f20 · outbound

This paper cites and Yuan, Y.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Yuan, Y

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.585508Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c23494a3-6110-4f42-b1fb-ef0b3e02ddac · outbound

This paper cites Scalable Second Order Optimization for Deep Learning.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Scalable Second Order Optimization for Deep Learning

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.591712Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 489e2f04-193e-4726-9e4f-70c67a967204 · outbound

This paper cites C., Foster, D.

MARS: Unleashing the Power of Variance Reduction for Training Large Models C., Foster, D

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.597515Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a3cade9e-05c8-4b36-a45c-aa930683b285 · outbound

This paper cites and Glynn, P.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Glynn, P

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.603039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 75163b56-d8a4-4c74-b8e2-2ff5c75ab8cd · outbound

This paper cites Old Optimizer, New Norm: An Anthology.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Old Optimizer, New Norm: An Anthology

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.608492Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 34e7797b-eb8b-4859-a7d8-4b4e61f5733d · outbound

This paper cites L., Gao, J., and Choi, Y.

MARS: Unleashing the Power of Variance Reduction for Training Large Models L., Gao, J., and Choi, Y

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.615584Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0af577b4-de4b-447b-a9c0-132775e4e360 · outbound

This paper cites Language Models are Few-Shot Learners.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Language Models are Few-Shot Learners

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.622016Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18cd5d9d-fd4e-4d79-9728-7799a3146168 · outbound

This paper cites Stochastic spectral descent for restricted boltzmann machines.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Stochastic spectral descent for restricted boltzmann machines

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.629196Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f271a367-fd5f-4c90-be6b-be2c2a5d4c4f · outbound

This paper cites Stochastic spectral descent for discrete graphical models.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Stochastic spectral descent for discrete graphical models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.635158Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6a327634-7f86-4bdd-abcb-41771812372d · outbound

This paper cites Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.642149Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5a69e144-f8f0-4d1e-a106-01c139073854 · outbound

This paper cites Symbolic discovery of optimization algorithms.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Symbolic discovery of optimization algorithms

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.647714Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation de0da958-9185-4240-ab0b-e12a0a20c986 · outbound

This paper cites W., Sutton, C., Gehrmann, S., et al.

MARS: Unleashing the Power of Variance Reduction for Training Large Models W., Sutton, C., Gehrmann, S., et al

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.652838Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4dfdbe6d-757e-4d09-aded-a15b2248b3ce · outbound

This paper cites Boolq: Exploring the surprising difficulty of natural yes/no questions.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Boolq: Exploring the surprising difficulty of natural yes/no questions

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.657716Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6986463-4542-4b2e-9862-06881e5d84fe · outbound

This paper cites and Orabona, F.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Orabona, F

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.662494Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11856636-9353-4a77-a17b-83886cbbd8a4 · outbound

This paper cites and Bottou, L.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Bottou, L

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.667415Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 383cc93d-6e19-4562-b8f7-bf25784f1e2f · outbound

This paper cites Saga: A fast incremental gradient method with support for non-strongly convex composite objectives.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Saga: A fast incremental gradient method with support for non-strongly convex composite objectives

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.672447Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 473f2f71-6b58-4f5c-baaf-687600880774 · outbound

This paper cites The Road Less Scheduled.

MARS: Unleashing the Power of Variance Reduction for Training Large Models The Road Less Scheduled

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.677433Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 311715d3-4e4b-4421-a965-f5f1ccda8f22 · outbound

This paper cites Stochastic variance-reduced newton: Accelerating finite-sum minimization with large batches.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Stochastic variance-reduced newton: Accelerating finite-sum minimization with large batches

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.683546Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a9ce96ff-9c55-4f14-b1e6-4d48be2963c5 · outbound

This paper cites BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.

MARS: Unleashing the Power of Variance Reduction for Training Large Models BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.688005Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c084cd7b-690e-4278-8d2c-dd4f5fcfbf75 · outbound

This paper cites The Llama 3 Herd of Models.

MARS: Unleashing the Power of Variance Reduction for Training Large Models The Llama 3 Herd of Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.692307Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 82f877d8-2988-4dd1-9e2e-4e596f507325 · outbound

This paper cites Adaptive subgradient methods for online learning and stochastic optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Adaptive subgradient methods for online learning and stochastic optimization

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.696550Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7450097-b671-4ddc-96e0-bb465215a037 · outbound

This paper cites J., Lin, Z., and Zhang, T.

MARS: Unleashing the Power of Variance Reduction for Training Large Models J., Lin, Z., and Zhang, T

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.701454Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d2867a9b-db15-4931-a3b8-dbaa05ed9173 · outbound

This paper cites Promise: Preconditioned stochastic optimization methods by incorporating scalable curvature estimates.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Promise: Preconditioned stochastic optimization methods by incorporating scalable curvature estimates

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.977171Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 28f611b6-5a1a-40d5-8130-e98ebebb5e23 · outbound

This paper cites A framework for few-shot language model evaluation, 07 2024.

MARS: Unleashing the Power of Variance Reduction for Training Large Models A framework for few-shot language model evaluation, 07 2024

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.711300Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 332c3eeb-a786-460f-91d7-5586729a2c81 · outbound

This paper cites and Lan, G.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Lan, G

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.715934Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0a9c5b38-f97a-4ac3-ad29-1b5a50e3f11e · outbound

This paper cites Openwebtext corpus.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Openwebtext corpus

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.720780Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0090491-0933-4f80-86d8-14c98d7ad1e8 · outbound

This paper cites and Graves, A.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Graves, A

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.725724Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c99be9de-a6d6-4e29-ab8a-075002b83bb4 · outbound

This paper cites Shampoo: Preconditioned stochastic tensor optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Shampoo: Preconditioned stochastic tensor optimization

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.730667Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 837873a5-4868-487b-9676-6051d1199ade · outbound

This paper cites Beyond convexity: Stochastic quasi-convex optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Beyond convexity: Stochastic quasi-convex optimization

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.735999Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0eb16ae7-421a-4793-818f-b9012f330f17 · outbound

This paper cites Deep residual learning for image recognition.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Deep residual learning for image recognition

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.740817Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 365be778-5806-4500-81d9-2650b84c20ee · outbound

This paper cites Measuring massive multitask language understanding.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Measuring massive multitask language understanding

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.863789Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 7c0c1a4f-42c9-4135-80fe-d10eab8c40bc · outbound

This paper cites an unresolved cited work.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Unresolved cited work

Reference 32

Resolution: unresolved; raw_fallback, observed 2026-08-12T19:43:03.843349Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation a44fc610-5fee-424c-903f-eeff8f848111 · outbound

This paper cites MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies.

MARS: Unleashing the Power of Variance Reduction for Training Large Models MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.756402Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f860c1b5-000b-43e3-90d6-ad060a880bd7 · outbound

This paper cites Super-adam: faster and universal framework of adaptive gradients.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Super-adam: faster and universal framework of adaptive gradients

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.815290Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 095ffd0c-484f-441a-958f-0ccf4ac6f5a4 · outbound

This paper cites and Zhang, T.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Zhang, T

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.766979Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6ac791e-5db5-4467-b101-9f19dc00ae43 · outbound

This paper cites Muon: An optimizer for hidden layers in neural networks, 2024.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Muon: An optimizer for hidden layers in neural networks, 2024

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.771826Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22458ebd-08da-48ea-9dec-10abb76e7a98 · outbound

This paper cites an unresolved cited work.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Unresolved cited work

Reference 37

Resolution: unresolved; raw_fallback, observed 2026-08-12T19:43:03.761712Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation d99fb412-a9cf-421e-93ec-20e084cb9f06 · outbound

This paper cites an unresolved cited work.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Unresolved cited work

Reference 38

Resolution: unresolved; raw_fallback, observed 2026-08-12T19:43:03.735491Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 7db134d0-4a1a-44a0-8ce9-914866c1da56 · outbound

This paper cites T., and Cevher, V.

MARS: Unleashing the Power of Variance Reduction for Training Large Models T., and Cevher, V

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.699534Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation e0978da1-f323-430f-9617-c1a31cc3e799 · outbound

This paper cites an unresolved cited work.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Unresolved cited work

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.791022Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 579e6a5e-ef91-48af-ae00-7b31d2e79891 · outbound

This paper cites Learning multiple layers of features from tiny images.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Learning multiple layers of features from tiny images

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.795890Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cce1d6db-0167-4dcb-b47f-59bb14ca9253 · outbound

This paper cites On the computation of the matrix k-th root.

MARS: Unleashing the Power of Variance Reduction for Training Large Models On the computation of the matrix k-th root

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.646413Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 6b6aa78d-cce2-4b06-9429-5f6a6637dc43 · outbound

This paper cites S., Moeller, T.

MARS: Unleashing the Power of Variance Reduction for Training Large Models S., Moeller, T

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.621762Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 564e5370-2bc4-40a4-bf8c-75b95e12b00b · outbound

This paper cites Gradient-based learning applied to document recognition.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Gradient-based learning applied to document recognition

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.813004Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29b4ed67-9353-458f-8251-93feac329703 · outbound

This paper cites Storm+: Fully adaptive sgd with recursive momentum for nonconvex optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Storm+: Fully adaptive sgd with recursive momentum for nonconvex optimization

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.583059Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 49389a72-448f-4e61-b719-5bd1f2b428dc · outbound

This paper cites Smoothness and Adaptivity in Nonlinear Optimization for Machine Learning Applications.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Smoothness and Adaptivity in Nonlinear Optimization for Machine Learning Applications

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.562677Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 28f3b14e-136d-4d1c-a3ae-94871d9d4cae · outbound

This paper cites Convergence of adam under relaxed assumptions.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Convergence of adam under relaxed assumptions

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.541573Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 476a60c7-8422-4d96-af61-7dabfc07b7ce · outbound

This paper cites DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model.

MARS: Unleashing the Power of Variance Reduction for Training Large Models DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.833269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6704f74b-430b-4ac6-aabf-bd407257fa08 · outbound

This paper cites Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.838625Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 929c6321-4e29-4615-9ed1-0bff2eb4f95e · outbound

This paper cites Adam$^+$: A Stochastic Method with Adaptive Variance Reduction.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Adam$^+$: A Stochastic Method with Adaptive Variance Reduction

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.844215Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98bfc228-49f1-4c1d-90ea-7fcf22fbdc3f · outbound

This paper cites and Hutter, F.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Hutter, F

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.849372Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7feaa9b5-925b-4996-a482-6bb4a927d258 · outbound

This paper cites Fineweb-edu: the finest collection of educational content, 2024.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Fineweb-edu: the finest collection of educational content, 2024

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.498210Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 0d21d3e8-2f8b-4f1a-b690-c0ab786f0e98 · outbound

This paper cites and Grosse, R.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Grosse, R

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.860686Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c7297947-5172-401e-93a6-a3807bcf4d5b · outbound

This paper cites and Tropp, J.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Tropp, J

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.453737Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation c3cd4b70-9543-4b57-8d1f-e0488227881d · outbound

This paper cites An Empirical Model of Large-Batch Training.

MARS: Unleashing the Power of Variance Reduction for Training Large Models An Empirical Model of Large-Batch Training

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.872324Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2c25452-1b46-4637-9aff-884ce3bff0e5 · outbound

This paper cites Adaptive Bound Optimization for Online Convex Optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Adaptive Bound Optimization for Online Convex Optimization

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.879308Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 68ea52d1-0b27-4be9-be39-b1dd82794446 · outbound

This paper cites Can a suit of armor conduct electricity? A new dataset for open book question answering.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Can a suit of armor conduct electricity? A new dataset for open book question answering

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.885057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation af18d8a0-4378-440b-9257-b35ebf83d8c2 · outbound

This paper cites A New Perspective on Shampoo's Preconditioner.

MARS: Unleashing the Power of Variance Reduction for Training Large Models A New Perspective on Shampoo's Preconditioner

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.891022Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 85cfd310-4490-42d3-a135-aa6de65968f7 · outbound

This paper cites A method for solving the convex programming problem with convergence rate o(1/k^2).

MARS: Unleashing the Power of Variance Reduction for Training Large Models A method for solving the convex programming problem with convergence rate o(1/k^2)

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.898101Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2845d882-48fc-4566-8ce3-ed4e2923103f · outbound

This paper cites Introductory lectures on convex optimization: A basic course, volume 87.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Introductory lectures on convex optimization: A basic course, volume 87

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.410000Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4ff75ab9-714f-4a67-b7f3-503bb0df5637 · outbound

This paper cites M., Liu, J., Scheinberg, K., and Tak \'a c , M.

MARS: Unleashing the Power of Variance Reduction for Training Large Models M., Liu, J., Scheinberg, K., and Tak \'a c , M

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.379053Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3f2a9ec6-1ccb-468c-bb19-2d2e95021808 · outbound

This paper cites Stochastic Recursive Gradient Algorithm for Nonconvex Optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Stochastic Recursive Gradient Algorithm for Nonconvex Optimization

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.913923Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 16aaf30a-2629-4f06-990b-25a4de6a3f11 · outbound

This paper cites Language models are unsupervised multitask learners.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Language models are unsupervised multitask learners

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.919275Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28c67352-ff2d-457b-b4a0-f5a51d15074e · outbound

This paper cites J., Hefny, A., Sra, S., Poczos, B., and Smola, A.

MARS: Unleashing the Power of Variance Reduction for Training Large Models J., Hefny, A., Sra, S., Poczos, B., and Smola, A

Reference 64

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.332688Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 64e703b0-6003-4ae7-ac02-cbd442a56063 · outbound

This paper cites On the Convergence of Adam and Beyond.

MARS: Unleashing the Power of Variance Reduction for Training Large Models On the Convergence of Adam and Beyond

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.936920Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 43854652-eacd-442b-9a6d-4ecb8e5021ab · outbound

This paper cites and Braun, H.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Braun, H

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.306454Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation eea572f5-4d30-4393-b5c5-794629c95ca1 · outbound

This paper cites A stochastic gradient method with an exponential convergence \_rate for finite training sets.

MARS: Unleashing the Power of Variance Reduction for Training Large Models A stochastic gradient method with an exponential convergence \_rate for finite training sets

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.950051Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7709864d-3762-4722-a6f1-245e4196a9ce · outbound

This paper cites L., Bhagavatula, C., and Choi, Y.

MARS: Unleashing the Power of Variance Reduction for Training Large Models L., Bhagavatula, C., and Choi, Y

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.262329Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ed306893-a7be-47ee-b441-72d8d69ab381 · outbound

This paper cites English Conversational Telephone Speech Recognition by Humans and Machines.

MARS: Unleashing the Power of Variance Reduction for Training Large Models English Conversational Telephone Speech Recognition by Humans and Machines

Reference 70

Resolution: verified exact; local_arxiv, observed 2026-08-12T19:43:02.356237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation b4caae8a-a9b1-4331-a710-0b162f46e2bc · outbound

This paper cites and Smola, A.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Smola, A

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.966520Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 85de6cc9-7d30-49f6-baaa-a23a13a6ab46 · outbound

This paper cites Iterative berechnung der reziproken matrix.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Iterative berechnung der reziproken matrix

Reference 72

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.209999Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8bbaeb57-1417-41df-9d90-d1be52929e5c · outbound

This paper cites and Zhang, T.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Zhang, T

Reference 73

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.173456Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation c262343c-682c-4473-8583-183ee134a835 · outbound

This paper cites and Stern, M.

MARS: Unleashing the Power of Variance Reduction for Training Large Models and Stern, M

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.983245Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 840b886c-4ca4-4976-9d25-026b41b288c8 · outbound

This paper cites A Distributed Data-Parallel PyTorch Implementation of the Distributed Shampoo Optimizer for Training Neural Networks At-Scale.

MARS: Unleashing the Power of Variance Reduction for Training Large Models A Distributed Data-Parallel PyTorch Implementation of the Distributed Shampoo Optimizer for Training Neural Networks At-Scale

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.988790Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d974669-e5d9-4404-b931-c8068638f8c8 · outbound

This paper cites Dropout: a simple way to prevent neural networks from overfitting.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Dropout: a simple way to prevent neural networks from overfitting

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:01.993564Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cfa80dc0-fdc0-4942-8de4-99571633fcc4 · outbound

This paper cites Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.107413Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 0024312a-7550-4bd0-ada2-61fad6a32166 · outbound

This paper cites Attention is all you need.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Attention is all you need

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.005034Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4a752e3d-2adc-4526-a48a-4cfed46f2a60 · outbound

This paper cites SOAP: Improving and Stabilizing Shampoo using Adam.

MARS: Unleashing the Power of Variance Reduction for Training Large Models SOAP: Improving and Stabilizing Shampoo using Adam

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.010064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 012af07a-9e4d-4b3d-836a-77b90ec463da · outbound

This paper cites Spiderboost and momentum: Faster variance reduction algorithms.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Spiderboost and momentum: Faster variance reduction algorithms

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.075581Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation b3dc8bbf-c6f3-4315-8dad-6e039b3dea6a · outbound

This paper cites Adagrad stepsizes: Sharp convergence over nonconvex landscapes.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Adagrad stepsizes: Sharp convergence over nonconvex landscapes

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.023654Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be36d960-c952-4b93-986c-5cdb9aa09499 · outbound

This paper cites Adan: Adaptive nesterov momentum algorithm for faster optimizing deep models.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Adan: Adaptive nesterov momentum algorithm for faster optimizing deep models

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.032612Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf3a7227-59be-4ac9-a24e-864cf24dc490 · outbound

This paper cites Quick and (not so) dirty: Unsupervised selection of justification sentences for multi-hop question answering.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Quick and (not so) dirty: Unsupervised selection of justification sentences for multi-hop question answering

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.039030Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08c21778-335f-47a5-9962-26cecaa098da · outbound

This paper cites A Coefficient Makes SVRG Effective.

MARS: Unleashing the Power of Variance Reduction for Training Large Models A Coefficient Makes SVRG Effective

Reference 84

Resolution: verified exact; local_arxiv, observed 2026-08-12T19:43:02.285588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation a3c723bf-65ea-48d1-8cee-0424688e03d4 · outbound

This paper cites Large Batch Optimization for Deep Learning: Training BERT in 76 minutes.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Large Batch Optimization for Deep Learning: Training BERT in 76 minutes

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.050371Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d6dde87-279b-4b4a-b5b7-79bfaf0dd268 · outbound

This paper cites ADADELTA: An Adaptive Learning Rate Method.

MARS: Unleashing the Power of Variance Reduction for Training Large Models ADADELTA: An Adaptive Learning Rate Method

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.057300Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11b4d049-9e91-4b71-bc3f-4cd504d7f7fc · outbound

This paper cites Hellaswag: Can a machine really finish your sentence? In Korhonen, A., Traum, D.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Hellaswag: Can a machine really finish your sentence? In Korhonen, A., Traum, D

Reference 87

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.063883Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2fd7335a-5e17-4e65-8923-dc475745f55a · outbound

This paper cites OPT: Open Pre-trained Transformer Language Models.

MARS: Unleashing the Power of Variance Reduction for Training Large Models OPT: Open Pre-trained Transformer Language Models

Reference 88

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.069466Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 107f6011-4904-4dc7-b21d-236b708babfd · outbound

This paper cites Adam can converge without any modification on update rules.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Adam can converge without any modification on update rules

Reference 89

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:03.004859Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation c62ad04f-5834-4868-a3e1-459c08c6a12a · outbound

This paper cites Deconstructing What Makes a Good Optimizer for Language Models.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Deconstructing What Makes a Good Optimizer for Language Models

Reference 90

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.080704Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e1daf55-ba58-4897-a71d-419a408daae3 · outbound

This paper cites Stochastic nested variance reduction for nonconvex optimization.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Stochastic nested variance reduction for nonconvex optimization

Reference 91

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:02.984590Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 54a450a1-df44-4fc5-b952-58b06cdd5cb6 · outbound

This paper cites Towards understanding convergence and generalization of adamw.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Towards understanding convergence and generalization of adamw

Reference 92

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T19:43:02.960616Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5465fbce-eb1c-4d54-a499-e15f24becaf1 · outbound

This paper cites C., Dvornek, N., Papademetris, X., and Duncan, J.

MARS: Unleashing the Power of Variance Reduction for Training Large Models C., Dvornek, N., Papademetris, X., and Duncan, J

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.099002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad9cbca0-2e2f-4589-ae04-316cdc5154a9 · outbound

This paper cites @esa (Ref.

MARS: Unleashing the Power of Variance Reduction for Training Large Models @esa (Ref

Reference 94

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.104287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 209059a6-ecd1-410f-926f-bbbd5bc28d06 · outbound

This paper cites an unresolved cited work.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Unresolved cited work

Reference 95

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.110478Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bafc3be5-4fca-4f01-9a61-c7cbb5b25d94 · outbound

This paper cites an unresolved cited work.

MARS: Unleashing the Power of Variance Reduction for Training Large Models Unresolved cited work

Reference 96

Resolution: unresolved; no resolver link, observed 2026-08-12T19:43:02.115828Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 82111d87-d0d8-4a25-a1e8-c4c8250ce563 · inbound

Physics of Skill Learning cites this paper.

Physics of Skill Learning MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-10T17:20:51.021889Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5c46eee-9c29-44dd-865b-18abd9153a49 · inbound

Training Deep Learning Models with Norm-Constrained LMOs cites this paper.

Training Deep Learning Models with Norm-Constrained LMOs MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 218

Resolution: verified exact; arxiv_id, observed 2026-05-21T21:22:37.163437Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 073a7acf-1bb7-471c-953e-9a2431184c4a · inbound

Continuous Cardiac Arrest Prediction in ICU using PPG Foundation Model cites this paper.

Continuous Cardiac Arrest Prediction in ICU using PPG Foundation Model MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-08T04:32:13.501247Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a2b38aa2-0c1f-4d77-81d3-f81653fca415 · inbound

Lions and Muons: Optimization via Stochastic Frank-Wolfe under Heavy-Tailed Noise cites this paper.

Lions and Muons: Optimization via Stochastic Frank-Wolfe under Heavy-Tailed Noise MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-07T10:58:37.720082Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a11c4c39-2f8d-4a92-a779-1caf95de60bb · inbound

DNT: a Deeply Normalized Transformer that can be trained by Momentum SGD cites this paper.

DNT: a Deeply Normalized Transformer that can be trained by Momentum SGD MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-06T14:54:58.857286Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 96305ccd-7657-4bf1-b514-55b65d9d72a7 · inbound

On the Convergence of Muon and Beyond cites this paper.

On the Convergence of Muon and Beyond MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 52

Resolution: verified exact; arxiv_id, observed 2026-05-18T15:56:34.038686Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 837ee271-41b3-4b54-aa56-2611a1f51e82 · inbound

Adaptive Memory Momentum via a Model-Based Framework for Deep Learning Optimization cites this paper.

Adaptive Memory Momentum via a Model-Based Framework for Deep Learning Optimization MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 71

Resolution: verified exact; arxiv_id, observed 2026-05-18T09:46:12.512191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 9d406b4a-3007-4676-bb2c-6be5b7626b11 · inbound

Optimal Sample Complexity for Single Time-Scale Actor-Critic with Momentum cites this paper.

Optimal Sample Complexity for Single Time-Scale Actor-Critic with Momentum MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 62

Resolution: verified exact; arxiv_id, observed 2026-05-16T08:17:36.532381Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation a767a7f1-1c97-42be-904b-2253a4f48c5c · inbound

HTMuon: Improving Muon via Heavy-Tailed Spectral Correction cites this paper.

HTMuon: Improving Muon via Heavy-Tailed Spectral Correction MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-05-25T06:55:26.336736Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8f4fb87c-5a45-480f-a4f1-3e547fba4e02 · inbound

CLion: Efficient Cautious Lion Optimizer with Enhanced Generalization cites this paper.

CLion: Efficient Cautious Lion Optimizer with Enhanced Generalization MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 34

Resolution: verified exact; arxiv_id, observed 2026-05-10T11:55:20.585950Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 020babd5-7115-47a9-b824-8019276a58f3 · inbound

Personalized Federated Learning for Gradient Alignment cites this paper.

Personalized Federated Learning for Gradient Alignment MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-09T05:50:26.268022Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 538e25c1-2441-4600-8506-41ad000ad01b · inbound

Optimizer-Model Consistency: Full Finetuning with the Same Optimizer as Pretraining Forgets Less cites this paper.

Optimizer-Model Consistency: Full Finetuning with the Same Optimizer as Pretraining Forgets Less MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 42

Resolution: verified exact; arxiv_id, observed 2026-05-11T19:26:08.551157Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 1c50ea8b-05e4-4edd-a635-6d520343531d · inbound

Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers cites this paper.

Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 165

Resolution: verified exact; arxiv_id, observed 2026-05-20T09:38:11.033711Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation e4e9ed29-c096-40da-a023-b20670e9989e · inbound

Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers cites this paper.

Symmetry-Compatible Principle for Optimizer Design: Embeddings, LM Heads, SwiGLU MLPs, and MoE Routers MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 168

Resolution: verified exact; arxiv_id, observed 2026-06-30T18:45:00.292382Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8c1a22ae-c71f-4321-af06-e8f23ac2b9cf · inbound

Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics cites this paper.

Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 158

Resolution: verified exact; arxiv_id, observed 2026-05-22T08:11:17.303623Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation b4b54b21-6276-4641-8006-bcd108a796c8 · inbound

MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization cites this paper.

MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-07-03T20:18:56.220618Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 72e002c3-65e5-4a39-875c-59bb0765fa72 · inbound

One-Step Gradient Delay is Not a Barrier for Large-Scale Asynchronous Pipeline Parallel LLM Pretraining cites this paper.

One-Step Gradient Delay is Not a Barrier for Large-Scale Asynchronous Pipeline Parallel LLM Pretraining MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 36

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:44:18.839572Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 92f28a8e-159e-4574-89fd-27f8a1ea7be4 · inbound

OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers cites this paper.

OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 128

Resolution: unresolved; no resolver link, observed 2026-07-11T22:10:49.683444Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd0f53ac-6b1c-47ec-96bf-825f239333de · inbound

Muse: Representation Geometry of Muon Beyond Normalized Momentum cites this paper.

Muse: Representation Geometry of Muon Beyond Normalized Momentum MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-02T01:58:58.664681Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f32997d-8304-4d9a-9fa0-e4c2397f32b3 · inbound

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining cites this paper.

OPTScientist: Multi-Agent Discovery of Typed Optimizer Programs for Transformer Pretraining MARS: Unleashing the Power of Variance Reduction for Training Large Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T12:30:38.237293Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.