Paper Citation Record · LEDGER

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving

As of 8 August 2026, this Paper Citation Record lists 44 of 44 outbound references and 0 inbound Pith citation observations for arXiv:2607.02640.

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

pith.paper-citation-record.v1
2607.02640 v1

Coverage vector

measured 44 of 44 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-12T08:09:44.957032Z

measured 44 of 44 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-07T06:34:17.273281+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

44 of 44 outbound references displayed

verified exact0
verified fuzzy0
unresolved44
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 7913e061-c235-49ec-a95c-d165f5445ffa · outbound

This paper cites Gulavani, Alexey Tumanov, and Ramachandran Ramjee.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Gulavani, Alexey Tumanov, and Ramachandran Ramjee

Reference 1

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1d694c90-0c10-4a9b-a911-4c83cad768ef · outbound

This paper cites GQA : Training generalized multi-query transformer models from multi-head checkpoints.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving GQA : Training generalized multi-query transformer models from multi-head checkpoints

Reference 2

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation caa99582-b910-437d-8549-33eace90fac1 · outbound

This paper cites Longformer: The Long-Document Transformer.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Longformer: The Long-Document Transformer

Reference 3

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 792b9d43-268f-415e-8a09-e3719e02c30c · outbound

This paper cites AudioLM: a Language Modeling Approach to Audio Generation.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving AudioLM: a Language Modeling Approach to Audio Generation

Reference 4

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0391e201-4bab-4d93-9251-afcf6f49c652 · outbound

This paper cites Generating Long Sequences with Sparse Transformers.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Generating Long Sequences with Sparse Transformers

Reference 5

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3156e588-8371-4414-acee-72c9438c2c2e · outbound

This paper cites Analysis of the increase and decrease algorithms for congestion avoidance in computer networks.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Analysis of the increase and decrease algorithms for congestion avoidance in computer networks

Reference 6

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d724d734-3542-47e8-aaf1-cd04c399785a · outbound

This paper cites Qwen2-Audio Technical Report.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Qwen2-Audio Technical Report

Reference 7

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8fc8e5dc-7d2f-42f3-8bea-81f1b1c3eed6 · outbound

This paper cites Franklin, Joseph E.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Franklin, Joseph E

Reference 8

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95b4d03c-cd20-4041-96b2-d22394729d2d · outbound

This paper cites Le, and Ruslan Salakhutdinov.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Le, and Ruslan Salakhutdinov

Reference 9

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6e70c58-2ffe-4c1f-90ef-40aaa8558757 · outbound

This paper cites FlashAttention-2 : Faster attention with better parallelism and work partitioning.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving FlashAttention-2 : Faster attention with better parallelism and work partitioning

Reference 10

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0349e6a-0555-486c-8fe0-cf3fdbedeca9 · outbound

This paper cites Fu, Stefano Ermon, Atri Rudra, and Christopher R \'e.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Fu, Stefano Ermon, Atri Rudra, and Christopher R \'e

Reference 11

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a57c4cb-18ea-47b3-b2a7-737322563186 · outbound

This paper cites High Fidelity Neural Audio Compression.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving High Fidelity Neural Audio Compression

Reference 12

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 38e28530-7865-46d3-9de8-6793765aaf2a · outbound

This paper cites Moshi: a speech-text foundation model for real-time dialogue.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Moshi: a speech-text foundation model for real-time dialogue

Reference 13

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6199ec6f-8f95-4bcb-a4d8-b464e14d29db · outbound

This paper cites LLaMA-Omni: Seamless Speech Interaction with Large Language Models.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving LLaMA-Omni: Seamless Speech Interaction with Large Language Models

Reference 14

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1626593a-ef9f-41e5-b1d4-2521593cda07 · outbound

This paper cites Serving DNNs like Clockwork : Performance predictability from the bottom up.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Serving DNNs like Clockwork : Performance predictability from the bottom up

Reference 15

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e712c485-e049-488f-b0b4-d51ccb89f2d2 · outbound

This paper cites Congestion avoidance and control.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Congestion avoidance and control

Reference 16

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20ec07b3-0f69-4411-a72e-482402927a55 · outbound

This paper cites Mistral 7B.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Mistral 7B

Reference 17

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d04d73b3-d5ae-49a0-8fb4-76c82a3a8e0b · outbound

This paper cites Efficient memory management for large language model serving with PagedAttention.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Efficient memory management for large language model serving with PagedAttention

Reference 18

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 79161ba7-16b2-41b2-8c9a-ee70a4163f74 · outbound

This paper cites SnapKV : LLM knows what you are looking for before generation.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving SnapKV : LLM knows what you are looking for before generation

Reference 19

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c59aa936-4a92-4502-95e7-90a772e5b03a · outbound

This paper cites AlpaServe : Statistical multiplexing with model parallelism for deep learning serving.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving AlpaServe : Statistical multiplexing with model parallelism for deep learning serving

Reference 20

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b8f498e0-88a6-48fa-8de5-e13712cdf3f2 · outbound

This paper cites an unresolved cited work.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Unresolved cited work

Reference 21

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 10693e2b-7197-42d1-b116-d5d2eff3dae6 · outbound

This paper cites Scissorhands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Scissorhands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time

Reference 22

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec53f533-d97c-4e4f-8954-5390443ac5a0 · outbound

This paper cites KIVI : A tuning-free asymmetric 2bit quantization for KV cache.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving KIVI : A tuning-free asymmetric 2bit quantization for KV cache

Reference 23

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3030f376-a360-4da2-9570-cf1ac3261d7a · outbound

This paper cites MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction

Reference 24

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 442a61c1-6515-4493-93d1-81e57c937a0e · outbound

This paper cites Splitwise: Efficient generative LLM inference using phase splitting.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Splitwise: Efficient generative LLM inference using phase splitting

Reference 25

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7a46b328-354d-481d-a5d2-c8f5da809c2a · outbound

This paper cites Pine AI : The most natural human-computer interface is your voice.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Pine AI : The most natural human-computer interface is your voice

Reference 26

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29cdf237-ab8f-4224-bf40-7f1dead3f717 · outbound

This paper cites Qwen3-Omni Technical Report.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Qwen3-Omni Technical Report

Reference 27

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab527647-3c8f-465d-a663-117925ee79a2 · outbound

This paper cites Robust speech recognition via large-scale weak supervision.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Robust speech recognition via large-scale weak supervision

Reference 28

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14668c51-e96d-4857-a407-6c7ef837539c · outbound

This paper cites Fast Transformer Decoding: One Write-Head is All You Need.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Fast Transformer Decoding: One Write-Head is All You Need

Reference 29

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 50d62345-6418-4ccf-a3f7-d4f1f7bf2756 · outbound

This paper cites FlexGen : High-throughput generative inference of large language models with a single GPU.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving FlexGen : High-throughput generative inference of large language models with a single GPU

Reference 30

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67b6f727-6870-44c9-b7e8-6baf1d842a3a · outbound

This paper cites SALMONN : Towards generic hearing abilities for large language models.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving SALMONN : Towards generic hearing abilities for large language models

Reference 31

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec92bf8f-d45a-4e8a-8f8f-3d59592af096 · outbound

This paper cites Interaction models: A scalable approach to human- AI collaboration.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Interaction models: A scalable approach to human- AI collaboration

Reference 32

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7ee98f8d-b19b-46b7-88f1-d3c53c725bff · outbound

This paper cites Gomez, ukasz Kaiser, and Illia Polosukhin.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Gomez, ukasz Kaiser, and Illia Polosukhin

Reference 33

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ea71684c-6268-479c-85b0-30887e9f7336 · outbound

This paper cites vLLM v0.23: Resumable requests and the realtime API.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving vLLM v0.23: Resumable requests and the realtime API

Reference 34

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f863a58-05dc-4328-aadb-78de0d1ae5ca · outbound

This paper cites Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers

Reference 35

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 88cd8c1d-2955-4039-939a-54a02b396b34 · outbound

This paper cites Efficient streaming language models with attention sinks.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Efficient streaming language models with attention sinks

Reference 36

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0539b802-1920-4ecb-864f-89a4875d842f · outbound

This paper cites Mini-Omni: Language Models Can Hear, Talk While Thinking in Streaming.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Mini-Omni: Language Models Can Hear, Talk While Thinking in Streaming

Reference 37

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a2ad763e-62d9-4a57-b2c3-abba3e4655c1 · outbound

This paper cites Qwen2.5-Omni Technical Report.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Qwen2.5-Omni Technical Report

Reference 38

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04baaf5c-23a6-453d-9144-222906d43bc4 · outbound

This paper cites FlashInfer : Efficient and customizable attention engine for LLM inference serving.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving FlashInfer : Efficient and customizable attention engine for LLM inference serving

Reference 39

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e7549eb-4052-408c-952e-be41a7ef4645 · outbound

This paper cites Orca: A distributed serving system for transformer-based generative models.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Orca: A distributed serving system for transformer-based generative models

Reference 40

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8789362b-f520-45a7-8901-7a5b49735e7a · outbound

This paper cites Big Bird : Transformers for longer sequences.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving Big Bird : Transformers for longer sequences

Reference 41

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 113ab0dd-8066-4480-8851-c5e224c9c8db · outbound

This paper cites H2O : Heavy-hitter oracle for efficient generative inference of large language models.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving H2O : Heavy-hitter oracle for efficient generative inference of large language models

Reference 42

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7cb5b5d8-948b-4a63-80e6-c26807a23ad1 · outbound

This paper cites SGLang : Efficient execution of structured language model programs.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving SGLang : Efficient execution of structured language model programs

Reference 43

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9af0105-998a-4824-b66c-51f86d8b97c4 · outbound

This paper cites DistServe : Disaggregating prefill and decoding for goodput-optimized large language model serving.

Metronome: Bound the Cache, Keep the Beat for Real-Time Interaction Model Serving DistServe : Disaggregating prefill and decoding for goodput-optimized large language model serving

Reference 44

Resolution: unresolved; no resolver link, observed 2026-07-12T08:09:44.957032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.