Paper Citation Record · LEDGER

Training LLMs with MXFP4

As of 15 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 22 inbound Pith citation observations for arXiv:2502.20586.

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

pith.paper-citation-record.v1
2502.20586 v3

Coverage vector

measured 0 of 0 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links

measured 22 of 22 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-15T06:32:42.880941+00:00

measured 22 of 22 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-15T21:03:33.260318Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-01T20:36:12.313960Z

Reference resolution

0 of 0 outbound references displayed

verified exact0
verified fuzzy0
unresolved0
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

No outbound reference observations are available for this paper version.

Pith citing papers

Observation 2135b4e7-81a8-42d8-8f41-5321d983ab34 · inbound

ProxSparse: Regularized Learning of Semi-Structured Sparsity Masks for Pretrained LLMs cites this paper.

ProxSparse: Regularized Learning of Semi-Structured Sparsity Masks for Pretrained LLMs Training LLMs with MXFP4

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-09T19:42:41.415287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3d4fff73-0d8f-464e-b762-f002f07774a4 · inbound

Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training cites this paper.

Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training Training LLMs with MXFP4

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-15T21:03:33.260318Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9c62804d-be7c-4241-939c-aaa8d16d30ba · inbound

Scaling Law for Quantization-Aware Training cites this paper.

Scaling Law for Quantization-Aware Training Training LLMs with MXFP4

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T15:41:08.862934Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30346e9c-fc01-46c8-b573-e6a49430264e · inbound

FP4 All the Way: Fully Quantized Training of LLMs cites this paper.

FP4 All the Way: Fully Quantized Training of LLMs Training LLMs with MXFP4

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T14:25:39.162969Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ad6dd0c-2c4d-450a-8a9c-19ee36aabc99 · inbound

Recipes for Pre-training LLMs with MXFP8 cites this paper.

Recipes for Pre-training LLMs with MXFP8 Training LLMs with MXFP4

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-07T12:12:47.436064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca415c67-07c8-4bb1-bf9b-d9649486e814 · inbound

OASIS: Outlier-Aware LUT-Based GEMM with Dual-Side Quantization for LLM Inference Acceleration cites this paper.

OASIS: Outlier-Aware LUT-Based GEMM with Dual-Side Quantization for LLM Inference Acceleration Training LLMs with MXFP4

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-06T11:15:34.031488Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0ee60707-4b4f-478a-ab91-f55567e38442 · inbound

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention cites this paper.

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention Training LLMs with MXFP4

Reference 28

Resolution: verified exact; arxiv_id, observed 2026-05-18T10:02:31.760017Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 72a1e432-e10a-4265-b36e-1250f8ef0ad8 · inbound

Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling cites this paper.

Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling Training LLMs with MXFP4

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-05-17T02:23:52.645493Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 25d502e3-7e94-4bf5-8b95-3035822adc04 · inbound

What is New in Stochastic Rounding: a Survey on Theory, Hardware, and Applications cites this paper.

What is New in Stochastic Rounding: a Survey on Theory, Hardware, and Applications Training LLMs with MXFP4

Reference 35

Resolution: malformed identifier; no resolver link, observed 2026-07-15T14:02:44.140957Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cbecd944-f006-4ae6-bb3b-6e4d659d0916 · inbound

VFA: Relieving Vector Operations in Flash Attention with Global Maximum Pre-computation cites this paper.

VFA: Relieving Vector Operations in Flash Attention with Global Maximum Pre-computation Training LLMs with MXFP4

Reference 24

Resolution: verified exact; arxiv_id, observed 2026-05-11T09:16:00.023031Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 49da8b73-31d6-4119-89de-76020a426fa0 · inbound

Grid Games: The Power of Multiple Grids for Quantizing Large Language Models cites this paper.

Grid Games: The Power of Multiple Grids for Quantizing Large Language Models Training LLMs with MXFP4

Reference 38

Resolution: verified exact; arxiv_id, observed 2026-05-13T06:47:26.520686Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 5e6d2583-f5b3-4f09-a78f-298ec611c463 · inbound

Search Your Block Floating Point Scales! cites this paper.

Search Your Block Floating Point Scales! Training LLMs with MXFP4

Reference 158

Resolution: verified exact; arxiv_id, observed 2026-05-13T06:02:24.073575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 54b05a9c-2c78-4252-b9c0-9351b7eab367 · inbound

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor cites this paper.

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor Training LLMs with MXFP4

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-05-21T07:59:50.002964Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 18e29319-1425-4b09-8442-593447c4cf42 · inbound

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor cites this paper.

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor Training LLMs with MXFP4

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-05-25T05:50:23.721565Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 8551d3bc-abaf-4612-b817-f9aff05f8454 · inbound

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor cites this paper.

Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor Training LLMs with MXFP4

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-06-30T18:04:57.989957Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 352a5745-22c9-4eb6-845c-1f2c47c8fed3 · inbound

Stochastic Rounding Increases Small Singular Values cites this paper.

Stochastic Rounding Increases Small Singular Values Training LLMs with MXFP4

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-07-01T20:26:13.686138Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 13654279-a9b7-47bd-aa9b-a45b319cb38b · inbound

Information-Theoretic Lower Bounds for Bit-Constrained Stochastic Optimization via a Reduction to Compressed Gaussian Mean Estimation cites this paper.

Information-Theoretic Lower Bounds for Bit-Constrained Stochastic Optimization via a Reduction to Compressed Gaussian Mean Estimation Training LLMs with MXFP4

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-07-01T20:36:12.315547Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 1738a3a2-3203-4f81-bbd8-8603df924636 · inbound

Full-Stack FP4: Stable LLM Pretraining with Quantized Projections, Optimizers, and Attention cites this paper.

Full-Stack FP4: Stable LLM Pretraining with Quantized Projections, Optimizers, and Attention Training LLMs with MXFP4

Reference 20

Resolution: unresolved; no resolver link, observed 2026-07-11T19:17:59.044982Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db2aac41-95f8-4dab-9921-d875adc75179 · inbound

Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection cites this paper.

Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection Training LLMs with MXFP4

Reference 26

Resolution: unresolved; no resolver link, observed 2026-07-14T15:32:26.691504Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7300d75f-2f49-4c8c-9904-e69dbccdb10b · inbound

CANN Bench: Benchmarking Agent Generated Kernels against Real NPU and Algorithmic Limits cites this paper.

CANN Bench: Benchmarking Agent Generated Kernels against Real NPU and Algorithmic Limits Training LLMs with MXFP4

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-02T08:11:51.965627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bbb63f90-a6d0-40e8-899d-8bc3351116a2 · inbound

Stable FP4 Training via Transposition-Invariant Block Quantization cites this paper.

Stable FP4 Training via Transposition-Invariant Block Quantization Training LLMs with MXFP4

Reference 17

Resolution: unresolved; no resolver link, observed 2026-07-31T05:04:00.386588Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f67e8c93-a6f1-4516-986c-ab1c8d5c4500 · inbound

HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models cites this paper.

HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models Training LLMs with MXFP4

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-01T14:11:20.981325Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.