Paper Citation Record · LEDGER

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points

As of 11 August 2026, this Paper Citation Record lists 43 of 43 outbound references and 0 inbound Pith citation observations for arXiv:2605.17471.

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

pith.paper-citation-record.v1
2605.17471 v1

Coverage vector

measured 43 of 43 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-20T14:53:22.233844Z

measured 43 of 43 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-11T06:34:44.6726+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

43 of 43 outbound references displayed

verified exact5
verified fuzzy29
unresolved9
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation d3737db0-9ab2-4d0b-a800-d78a18c765fb · outbound

This paper cites The potential of second-order optimization for llms: A study with full gauss-newton.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points The potential of second-order optimization for llms: A study with full gauss-newton

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.712095Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 93ec69c9-ff8c-40ce-bb3f-5088c1f10601 · outbound

This paper cites L., Li, B., Cameron, P., Jaggi, M., Alistarh, D., Hoefler, T., and Hensman, J.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points L., Li, B., Cameron, P., Jaggi, M., Alistarh, D., Hoefler, T., and Hensman, J

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.717968Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 2bcbb0b3-15a2-4eb5-8d59-fb8df4dcd2b8 · outbound

This paper cites Proxquant: Quantized neural networks via proximal operators.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Proxquant: Quantized neural networks via proximal operators

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.705709Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation db0552a5-9597-4924-8eb3-29b83d43b370 · outbound

This paper cites Some large-scale matrix computation problems.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Some large-scale matrix computation problems

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.716154Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation f9410d21-7d57-44e7-a8d5-56439c792447 · outbound

This paper cites Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:53:31.622415Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation e529556b-460f-4ffb-a5f5-455ee9f62d87 · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 6

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.716612Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 4bd8853e-f3d2-44ce-b776-073edded6b19 · outbound

This paper cites Analysis of stochastic lanczos quadrature for spectrum approximation.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Analysis of stochastic lanczos quadrature for spectrum approximation

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.712839Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 23723677-baa0-4f82-9b79-034cbaa264e2 · outbound

This paper cites Fast hadamard transform in cuda, with a pytorch interface.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Fast hadamard transform in cuda, with a pytorch interface

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.714268Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 26be97c6-4312-439c-922f-f04011e5331d · outbound

This paper cites Gq-vae: A gated quantized vae for learning variable length tokens.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Gq-vae: A gated quantized vae for learning variable length tokens

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:53:31.637428Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation da13262c-f8aa-40cf-a570-59b4b24af29a · outbound

This paper cites W., and Keutzer, K.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points W., and Keutzer, K

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.677044Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 9a9592b1-e487-487a-95bd-2deba3b2af6f · outbound

This paper cites W., and Keutzer, K.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points W., and Keutzer, K

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.694324Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 2f1a770a-a99c-4956-92c6-40745fa2248e · outbound

This paper cites K., McKinstry, J.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points K., McKinstry, J

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.698385Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 98c80ae9-ac0d-4f31-84f7-8767a9b8f03e · outbound

This paper cites Training with quantization noise for extreme model compression.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Training with quantization noise for extreme model compression

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.684709Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 6b3d8056-8292-4a16-a977-e3586964cdec · outbound

This paper cites G., Duan, D., Iyengar, A., Liu, J.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points G., Duan, D., Iyengar, A., Liu, J

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.681827Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation fd92ed3e-5cab-4969-9406-fe7252cdafd1 · outbound

This paper cites and Alistarh, D.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points and Alistarh, D

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.678018Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation f6a74c3c-9772-4459-8dea-e26fffb5c515 · outbound

This paper cites Gptq: Accurate post-training quantization for generative pre-trained transformers.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Gptq: Accurate post-training quantization for generative pre-trained transformers

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.690723Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 20a99394-9bd0-4199-8b24-4188705af4e5 · outbound

This paper cites W., and Keutzer, K.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points W., and Keutzer, K

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.702365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 42bc2a11-ebde-4ff1-bda4-4443b62cd60b · outbound

This paper cites An investigation into neural net optimization via hessian eigenvalue density.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points An investigation into neural net optimization via hessian eigenvalue density

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.700496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 81a5202f-292f-4bb6-a800-79f136f806dd · outbound

This paper cites Quantization and training of neural networks for efficient integer-arithmetic-only inference.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Quantization and training of neural networks for efficient integer-arithmetic-only inference

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.683608Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation c211b17f-fb15-465f-950b-bb9bc7e9ca19 · outbound

This paper cites M., and Jordan, M.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points M., and Jordan, M

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.688965Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 47458962-63a1-483f-bd88-35e710f8da6d · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 21

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.676237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation adbba272-da8f-4cfe-96cf-d419a15b895a · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 22

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.696164Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a7e92d09-50ce-4625-aef8-c2c504198620 · outbound

This paper cites F., Bordelon, B., Muennighoff, N., Paul, M., Pehlevan, C., Re, C., and Raghunathan, A.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points F., Bordelon, B., Muennighoff, N., Paul, M., Pehlevan, C., Re, C., and Raghunathan, A

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.649480Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation aafcbcbf-f7e7-43d7-8e91-5a912031b0c4 · outbound

This paper cites H., Gil, S., Anand, N., and Kakade, S.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points H., Gil, S., Anand, N., and Kakade, S

Reference 24

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:53:31.640359Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 997d89e7-2822-4192-a40a-1074f38dcd33 · outbound

This paper cites Awq: Activation-aware weight quantization for on-device llm compression and acceleration.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Awq: Activation-aware weight quantization for on-device llm compression and acceleration

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.680014Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation fee49741-bffb-4acb-aa68-44957e9f1f9d · outbound

This paper cites Bit: Robustly binarized multi-distilled transformer.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Bit: Robustly binarized multi-distilled transformer

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.682883Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 2464149b-950e-4173-8520-67d86a60c536 · outbound

This paper cites Spinquant: Llm quantization with learned rotations.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Spinquant: Llm quantization with learned rotations

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.692322Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation f2838259-fded-4492-87c7-d965dff96121 · outbound

This paper cites Paretoq: Scaling laws in extremely low-bit llm quantization.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Paretoq: Scaling laws in extremely low-bit llm quantization

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.686490Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation d22af3dc-2c0b-4c55-aae6-08ed2b3d32ec · outbound

This paper cites A., Datta, P., Dean, J., Jain, P., and Kusupati, A.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points A., Datta, P., Dean, J., Jain, P., and Kusupati, A

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.703805Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation eb481fb0-a97d-4fb1-84be-50d78f135a71 · outbound

This paper cites Lectures on convex optimization, volume 137.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Lectures on convex optimization, volume 137

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.679021Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b7e065f4-8a12-4866-acb2-927f8c8f00df · outbound

This paper cites L., Nikdan, M., and Alistarh, D.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points L., Nikdan, M., and Alistarh, D

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.687211Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation e096898f-08db-494a-badf-593a2e10c5ee · outbound

This paper cites CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points CAGE: Curvature-Aware Gradient Estimation For Accurate Quantization-Aware Training

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-06-19T17:11:53.086525Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b850bf47-6eed-4ef3-ae19-5290f27c210c · outbound

This paper cites BitNet: Scaling 1-bit Transformers for Large Language Models.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points BitNet: Scaling 1-bit Transformers for Large Language Models

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:53:31.637256Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 4fc41ef6-372e-48f3-b42e-67d7b1950451 · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 34

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.669174Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation d79b8b15-40f4-4a84-86a0-f60e04b35d34 · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 35

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.698090Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation f69b5a88-58e5-40d6-acb5-e8ac03b48245 · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 36

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.674991Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b0baf9b0-14ef-410e-b5fa-9050845afb61 · outbound

This paper cites Modulora: finetuning 2-bit llms on consumer gpus by integrating with modular quantizers.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Modulora: finetuning 2-bit llms on consumer gpus by integrating with modular quantizers

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.661389Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 4e903211-2259-46df-a47f-def832eee661 · outbound

This paper cites Large batch optimization for deep learning: Training bert in 76 minutes.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Large batch optimization for deep learning: Training bert in 76 minutes

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.701966Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation beaa2435-6e3d-456d-9a1e-2b745bbbbab3 · outbound

This paper cites R., Li, D., and Ju, H.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points R., Li, D., and Ju, H

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.699748Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation ed98a403-b67e-4295-baa3-6aed1fb24758 · outbound

This paper cites Q-hitter: A better token oracle for efficient llm inference via sparse-quantized kv cache.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Q-hitter: A better token oracle for efficient llm inference via sparse-quantized kv cache

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:53:31.696682Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b036e678-6d5b-49e9-9488-d0ebfa61e54b · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 41

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.694141Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 7a2edca3-e151-408f-824c-d184a3618838 · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 42

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.674276Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 54c375af-e964-417b-8fd1-9e9d8651aedb · outbound

This paper cites an unresolved cited work.

WinQ: Accelerating Quantization-Aware Training of Language Models Around Saddle Points Unresolved cited work

Reference 43

Resolution: unresolved; raw_fallback, observed 2026-05-20T14:53:31.685416Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Pith citing papers

No inbound Pith citation observations are available.