Paper Citation Record · LEDGER

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference

As of 9 August 2026, this Paper Citation Record lists 33 of 33 outbound references and 2 inbound Pith citation observations for arXiv:2507.19608.

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

pith.paper-citation-record.v1
2507.19608 v1

Coverage vector

measured 33 of 33 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T14:17:45.932407Z

measured 35 of 35 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-12T21:39:23.683913Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-11T09:05:57.878259Z

Reference resolution

33 of 33 outbound references displayed

verified exact1
verified fuzzy9
unresolved23
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 1bb866c0-c2ee-4039-aba0-55db7ac9af4f · outbound

This paper cites Sparks of artificial general intelligence: Early experiments with gpt-4,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Sparks of artificial general intelligence: Early experiments with gpt-4,

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:43.731989Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8cc926ef-9008-46d7-addb-30cbee7a21b6 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference LLaMA: Open and Efficient Foundation Language Models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:43.810078Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d75ec5c4-fa9e-4bc1-91ee-a62d1efc605f · outbound

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

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Training-Free Activation Sparsity in Large Language Models

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:43.859667Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c32676cb-7c12-4fdf-bee0-b0dcaa3a82c8 · outbound

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

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Awq: Activation-aware weight quanti- zation for on-device llm compression and acceleration,

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:43.943050Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 813f3b92-36c4-4764-957c-7f8f3e84f12e · outbound

This paper cites Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.003867Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98267ba7-8eeb-4aeb-9e78-24ce97bd8e69 · outbound

This paper cites Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.088718Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c439604b-53ee-472f-a14c-54ce0aaf32a7 · outbound

This paper cites A Simple and Effective Pruning Approach for Large Language Models.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference A Simple and Effective Pruning Approach for Large Language Models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.141906Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eff45d7b-c6ee-48fb-9095-99acb06129ed · outbound

This paper cites How Sparse Attention Approximates Exact Attention? Your Attention is Naturally $n^C$-Sparse.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference How Sparse Attention Approximates Exact Attention? Your Attention is Naturally $n^C$-Sparse

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.221622Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14b49207-9130-4dc1-aea8-f1372c7c2330 · outbound

This paper cites Efficient Streaming Language Models with Attention Sinks.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Efficient Streaming Language Models with Attention Sinks

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.287910Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ada1c4d4-38ac-4c43-a5f7-507ee13b7a3d · outbound

This paper cites MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.377690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9bc145ef-2ed5-4754-a6dd-b45804deb286 · outbound

This paper cites Snapkv: Llm knows what you are looking for before generation,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Snapkv: Llm knows what you are looking for before generation,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:48.483105Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 30b4013c-6b1c-455b-af57-49cd22ef554b · outbound

This paper cites InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.508093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4be6df4-9618-4fd4-9dac-81eac982d980 · outbound

This paper cites FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.573805Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 367c8ba8-133a-4555-a6d7-cfb04a05d4c5 · outbound

This paper cites Spargeattn: Accurate sparse attention accelerating any model infer- ence,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Spargeattn: Accurate sparse attention accelerating any model infer- ence,

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.642377Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5b057b0-42ae-4ab4-a1d3-1dac12a19455 · outbound

This paper cites Delta networks for optimized recurrent network computation,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Delta networks for optimized recurrent network computation,

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:48.278871Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 55a03858-4c28-423b-a521-4320bec3ba52 · outbound

This paper cites Spartus: A 9.4 top/s fpga-based lstm accelerator exploiting spatio-temporal sparsity,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Spartus: A 9.4 top/s fpga-based lstm accelerator exploiting spatio-temporal sparsity,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:47.977207Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f62d9bf5-0fa3-410c-be2a-fd0556b92eda · outbound

This paper cites Skip- convolutions for efficient video processing,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Skip- convolutions for efficient video processing,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:47.747998Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation eca9242f-b1d8-412e-9d67-789dd30e7bb5 · outbound

This paper cites Deltacnn: End-to-end cnn inference of sparse frame differences in videos,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Deltacnn: End-to-end cnn inference of sparse frame differences in videos,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:47.556073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation c49cfb21-5b50-42b1-9c25-cc8284a9e67d · outbound

This paper cites Motiondeltacnn: Sparse cnn inference of frame differences in moving camera videos with spherical buffers and padded convolutions,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Motiondeltacnn: Sparse cnn inference of frame differences in moving camera videos with spherical buffers and padded convolutions,

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:44.949752Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c51d8b60-de41-47f5-9805-68571e4c02b0 · outbound

This paper cites Deltarnn: A power-efficient recurrent neural network accelerator,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Deltarnn: A power-efficient recurrent neural network accelerator,

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:47.341427Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation b7172d2b-2e33-45b2-8b26-2f3a2c9275ec · outbound

This paper cites Edge- drnn: Recurrent neural network accelerator for edge inference,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Edge- drnn: Recurrent neural network accelerator for edge inference,

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.081409Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40844b20-70c8-4430-a691-c182b1758ff7 · outbound

This paper cites Deltakws: A 65nm 36nj/decision bio-inspired temporal-sparsity- aware digital keyword spotting ic with 0.6v near-threshold sram,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Deltakws: A 65nm 36nj/decision bio-inspired temporal-sparsity- aware digital keyword spotting ic with 0.6v near-threshold sram,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:47.068334Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation cc23c3dc-0b9d-4de2-ad22-66d07e847c86 · outbound

This paper cites Delta Keyword Transformer: Bringing Transformers to the Edge through Dynamically Pruned Multi-Head Self-Attention.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Delta Keyword Transformer: Bringing Transformers to the Edge through Dynamically Pruned Multi-Head Self-Attention

Reference 23

Resolution: verified exact; local_arxiv, observed 2026-08-06T14:17:46.233450Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5e4ed24d-5b81-4f0e-acd8-ba66a82025a3 · outbound

This paper cites The Llama 3 Herd of Models.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference The Llama 3 Herd of Models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.325736Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ae4f796b-ec81-45e0-93e2-938f5161f25c · outbound

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

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference BitNet: Scaling 1-bit Transformers for Large Language Models

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.429662Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 28fd27a4-956e-4cff-a18c-1fa7239c453d · outbound

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

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.474224Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3581d286-bce0-44e9-8ed7-2d2ae873e068 · outbound

This paper cites BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.546320Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 49cb55d8-bda6-40f6-a0b1-c32ffd090bb9 · outbound

This paper cites HellaSwag: Can a Machine Really Finish Your Sentence?.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference HellaSwag: Can a Machine Really Finish Your Sentence?

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.627275Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f64321e-0319-4ede-a144-45813236f438 · outbound

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

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Can a suit of armor conduct electricity? a new dataset for open book question answering,

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.719990Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 278362cf-abed-460a-9bd9-e00c8c3a358e · outbound

This paper cites PIQA: Reasoning about Physical Commonsense in Natural Language.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference PIQA: Reasoning about Physical Commonsense in Natural Language

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.725322Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab7519b4-282c-47de-be97-862d5cb3dbc2 · outbound

This paper cites Winogrande: An adversarial winograd schema challenge at scale,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Winogrande: An adversarial winograd schema challenge at scale,

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:46.916485Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 303632bb-10d2-4d3e-8cae-a53c5d0e6e73 · outbound

This paper cites Know what you don’t know: Unanswerable questions for squad,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference Know what you don’t know: Unanswerable questions for squad,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T14:17:46.666514Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 53a69183-1a9b-475f-9758-b172a54593b4 · outbound

This paper cites The language model evaluation harness,.

DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference The language model evaluation harness,

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-06T14:17:45.932407Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation b9c372f8-f66e-4582-b2d8-0b483a5920f6 · inbound

Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation cites this paper.

Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference

Reference 72

Resolution: verified exact; arxiv_id, observed 2026-05-11T09:05:57.881229Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation e90559e1-f94b-41f0-b376-5abd55d0a4de · inbound

AEyeDE: An Attention-Based Attribution Framework for AI-Generated Text Detection cites this paper.

AEyeDE: An Attention-Based Attribution Framework for AI-Generated Text Detection DeltaLLM: A Training-Free Framework Exploiting Temporal Sparsity for Efficient Edge LLM Inference

Reference 32

Resolution: unresolved; no resolver link, observed 2026-07-12T21:39:23.683913Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.