Paper Citation Record · LEDGER

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches

As of 6 August 2026, this Paper Citation Record lists 29 of 29 outbound references and 0 inbound Pith citation observations for arXiv:2509.22166.

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

pith.paper-citation-record.v1
2509.22166 v4

Coverage vector

measured 29 of 29 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-18T13:36:55.938673Z

measured 29 of 29 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-05T06:32:48.257954+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

29 of 29 outbound references displayed

verified exact19
verified fuzzy4
unresolved1
parse uncertain0
malformed identifier2
metadata mismatch3

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation e816b420-7e09-4425-98d7-ba16d22ca3e6 · outbound

This paper cites Amber Pruner: Leveraging N:M Activation Sparsity for Efficient Prefill in Large Language Models.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Amber Pruner: Leveraging N:M Activation Sparsity for Efficient Prefill in Large Language Models

Reference 1

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.922214Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation dda4dc23-5b3d-4cf9-9abb-97e25ee1380e · outbound

This paper cites Post-Training Statistical Calibration for Higher Activation Sparsity.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Post-Training Statistical Calibration for Higher Activation Sparsity

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.932269Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 62b638da-e925-4119-930d-be8c9dec6eab · outbound

This paper cites Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:41:25.942424Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 2c2d721a-ed0f-4054-8091-7fdcf7797b3f · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Training Verifiers to Solve Math Word Problems

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:41:25.958174Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 333bcaa5-066f-428e-9eee-ded1225fba65 · outbound

This paper cites Extreme Compression of Large Language Models via Additive Quantization.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Extreme Compression of Large Language Models via Additive Quantization

Reference 5

Resolution: metadata mismatch; arxiv_id, observed 2026-05-18T13:41:25.968867Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 2e1dac95-2868-46bb-aa47-e2a6dd506406 · outbound

This paper cites Inference economics of language models.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Inference economics of language models

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.963890Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 075d5130-5ac7-47cb-bf43-0b36bc1db935 · outbound

This paper cites GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:41:25.937127Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation fd974e0a-185c-4c99-92dd-1a175add0849 · outbound

This paper cites 10 September 2025 Song Han, Jeff Pool, John Tran, and William J.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches 10 September 2025 Song Han, Jeff Pool, John Tran, and William J

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T13:42:39.642164Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 01cfc3cd-a349-45ad-a260-b3b82b6379e4 · outbound

This paper cites Learning both Weights and Connections for Efficient Neural Networks.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Learning both Weights and Connections for Efficient Neural Networks

Reference 9

Resolution: metadata mismatch; local_arxiv, observed 2026-05-18T13:41:25.927014Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation b72ef6f3-d102-4d7d-a340-e3e26eea07c0 · outbound

This paper cites Measuring Massive Multitask Language Understanding.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Measuring Massive Multitask Language Understanding

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:41:25.947610Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation b512806d-1f0f-4c30-9fdb-84921885c9b3 · outbound

This paper cites Accelerating Transformer Pre-training with 2:4 Sparsity.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Accelerating Transformer Pre-training with 2:4 Sparsity

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.953148Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 65f310cf-837c-40c2-a8ae-7c7b60e7477c · outbound

This paper cites R-Sparse R-CNN: SAR Ship Detection Based on Background-Aware Sparse Learnable Proposals.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches R-Sparse R-CNN: SAR Ship Detection Based on Background-Aware Sparse Learnable Proposals

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.896040Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 42a54f96-0d72-48cb-8149-34954a2b14b5 · outbound

This paper cites CATS: Contextually-Aware Thresholding for Sparsity in Large Language Models.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches CATS: Contextually-Aware Thresholding for Sparsity in Large Language Models

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.906626Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 42a286d6-9825-4df1-84eb-66fd99bea4dd · outbound

This paper cites Training-Free Activation Sparsity in Large Language Models.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Training-Free Activation Sparsity in Large Language Models

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.916994Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 4bb9905d-9a27-433f-ad34-af194e18c0e8 · outbound

This paper cites From 2:4 to 8:16 sparsity patterns in LLMs for Outliers and Weights with Variance Correction.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches From 2:4 to 8:16 sparsity patterns in LLMs for Outliers and Weights with Variance Correction

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:41:25.882054Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 63ce8a56-3f83-4c7d-8636-48bae66434ea · outbound

This paper cites ACE: Exploring Activation Cosine Similarity and Variance for Accurate and Calibration-Efficient LLM Pruning.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches ACE: Exploring Activation Cosine Similarity and Variance for Accurate and Calibration-Efficient LLM Pruning

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.901495Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation ca45b6e4-f873-4409-b5d2-5e1678fa5859 · outbound

This paper cites ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.911858Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 37acd4d9-9507-4a63-9fe5-a5e14bd99b14 · outbound

This paper cites The LAMBADA dataset: Word prediction requiring a broad discourse context.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches The LAMBADA dataset: Word prediction requiring a broad discourse context

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:41:25.877488Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 31ff3aa9-8356-4491-844e-10b705eb829a · outbound

This paper cites Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Unmasking the Lottery Ticket Hypothesis: What's Encoded in a Winning Ticket's Mask?

Reference 19

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.867677Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 06391375-4c11-4b53-801d-6121852fa91b · outbound

This paper cites arXiv preprint arXiv:2505.14884 , year=.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches arXiv preprint arXiv:2505.14884 , year=

Reference 20

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.872413Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 61704377-9c21-40f3-9c14-5ce4ebc0627d · outbound

This paper cites Yixin Song, Zeyu Mi, Haotong Xie, and Haibo Chen.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Yixin Song, Zeyu Mi, Haotong Xie, and Haibo Chen

Reference 21

Resolution: malformed identifier; arxiv_id, observed 2026-05-18T13:41:25.886789Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 61ede6c2-ba32-4ba1-8217-129d279a9873 · outbound

This paper cites Q-Sparse: All Large Language Models can be Fully Sparsely-Activated.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Q-Sparse: All Large Language Models can be Fully Sparsely-Activated

Reference 22

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.857400Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 3bb102d2-c0d5-4828-8536-cd6d2b436c5d · outbound

This paper cites GIFT-SW: Gaussian noise Injected Fine-Tuning of Salient Weights for LLMs.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches GIFT-SW: Gaussian noise Injected Fine-Tuning of Salient Weights for LLMs

Reference 23

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:41:25.862900Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 1f45a8d7-83f1-42d2-aa2b-b8b08e2d160a · outbound

This paper cites Instruction-Following Evaluation for Large Language Models.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Instruction-Following Evaluation for Large Language Models

Reference 24

Resolution: metadata mismatch; local_arxiv, observed 2026-05-18T13:41:25.891407Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation aa2c5424-3631-4f47-adf9-f0d8c07afce4 · outbound

This paper cites an unresolved cited work.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Unresolved cited work

Reference 25

Resolution: unresolved; raw_fallback, observed 2026-05-18T13:42:39.635668Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 4ca36bec-ba95-47cb-b9e2-bec6a2fc6963 · outbound

This paper cites Dataset Description Metric WikiText-2 (Merity et al.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Dataset Description Metric WikiText-2 (Merity et al

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T13:42:39.628741Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 44be1378-79bf-4b66-81e1-26bde0b437bd · outbound

This paper cites Contains 5957 4-way multiple-choice questions.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Contains 5957 4-way multiple-choice questions

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T13:42:39.625675Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 7a1d85b8-5a7d-4ba9-9aef-81dcaeffaef7 · outbound

This paper cites Accuracy (Prompt-level) Accuracy (Instruct-level) 14 September 2025 D WEIGHTS VERSUSACTIVATIONS Table 9: The performance of models with applied unstructured activation pruning.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Accuracy (Prompt-level) Accuracy (Instruct-level) 14 September 2025 D WEIGHTS VERSUSACTIVATIONS Table 9: The performance of models with applied unstructured activation pruning

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-05-18T13:42:39.632367Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Observation 7cd3ccff-a35d-46da-9c3d-4c42e4552e82 · outbound

This paper cites an unresolved cited work.

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches Unresolved cited work

Reference 29

Resolution: malformed identifier; raw_fallback, observed 2026-05-18T13:42:39.638808Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-05T06:32:48.257954+00:00.

Pith citing papers

No inbound Pith citation observations are available.