Paper Citation Record · LEDGER

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory

As of 14 August 2026, this Paper Citation Record lists 26 of 26 outbound references and 0 inbound Pith citation observations for arXiv:2607.28263.

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

pith.paper-citation-record.v1
2607.28263 v1

Coverage vector

measured 26 of 26 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-31T12:56:46.806841Z

measured 26 of 26 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-14T06:32:32.682623+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

26 of 26 outbound references displayed

verified exact0
verified fuzzy0
unresolved25
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 5b257e32-d26d-40ff-9450-4edbb1cfdd63 · outbound

This paper cites The Llama 3 Herd of Models.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory The Llama 3 Herd of Models

Reference 4

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.612278Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c72022f2-797a-4c62-8851-a4c96ce608b8 · outbound

This paper cites RULER: What's the Real Context Size of Your Long-Context Language Models?.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory RULER: What's the Real Context Size of Your Long-Context Language Models?

Reference 6

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.892297Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14ff5225-b6c4-4a1c-948b-b7bfa0922871 · outbound

This paper cites RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation

Reference 7

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.993564Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf6cf282-fc70-4d4e-94e3-7cd519c5a8df · outbound

This paper cites The Remarkable Robustness of LLMs: Stages of Inference?.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory The Remarkable Robustness of LLMs: Stages of Inference?

Reference 8

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.090058Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd146709-5cf9-4f84-98d1-c9e0529c2a2f · outbound

This paper cites SnapKV: LLM Knows What You are Looking for Before Generation.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory SnapKV: LLM Knows What You are Looking for Before Generation

Reference 9

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.197106Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22bdc162-20b6-4d40-8af9-6cef3be9f150 · outbound

This paper cites CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

Reference 10

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.291959Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a10ddebc-b201-4562-84f4-d81c5d510e62 · outbound

This paper cites Evaluating Very Long-Term Conversational Memory of LLM Agents.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Evaluating Very Long-Term Conversational Memory of LLM Agents

Reference 11

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.432364Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d26e740b-b28b-4095-be6d-f5b487faa1c7 · outbound

This paper cites Landmark Attention: Random-Access Infinite Context Length for Transformers.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Landmark Attention: Random-Access Infinite Context Length for Transformers

Reference 12

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.516458Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 77902c20-8678-4c2e-9a89-59e4d2e6e36d · outbound

This paper cites YaRN: Efficient Context Window Extension of Large Language Models.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory YaRN: Efficient Context Window Extension of Large Language Models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.589864Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2166dd7d-26f5-4ead-8f8e-6888b228ecb0 · outbound

This paper cites arXiv preprint arXiv:2603.19664.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory arXiv preprint arXiv:2603.19664

Reference 14

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.699653Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb06958c-9289-4e84-b333-3ac60a66a525 · outbound

This paper cites You Only Cache Once: Decoder-Decoder Architectures for Language Models.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory You Only Cache Once: Decoder-Decoder Architectures for Language Models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.924478Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d377100a-6afc-4bf7-acc6-d42127968a84 · outbound

This paper cites Focused Transformer: Contrastive Training for Context Scaling.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Focused Transformer: Contrastive Training for Context Scaling

Reference 18

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.996091Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7642a1f2-9aac-4553-8f54-cffc0e58b46e · outbound

This paper cites MEMORYLLM: Towards Self-Updatable Large Language Models.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory MEMORYLLM: Towards Self-Updatable Large Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.177620Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc2c4eee-5961-41b9-800a-1342369c9d8b · outbound

This paper cites Layer-Condensed KV Cache for Efficient Inference of Large Language Models.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Layer-Condensed KV Cache for Efficient Inference of Large Language Models

Reference 21

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.292098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 246bac0e-0583-4083-ae8d-d7f5e39f296f · outbound

This paper cites CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion

Reference 23

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.484066Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56a026ec-e8b0-4b30-ada3-51ce465eab2a · outbound

This paper cites Long Context Compression with Activation Beacon.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Long Context Compression with Activation Beacon

Reference 24

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.586819Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56c6f3a7-c37f-45e0-ad71-8748f98dc2b8 · outbound

This paper cites InNeurIPS.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory InNeurIPS

Reference 25

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.698287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2053ddf8-536e-4843-82b6-62932083f24e · outbound

This paper cites †MemoryLLM uses a different chat-tuned backbone.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory †MemoryLLM uses a different chat-tuned backbone

Reference 92

Resolution: malformed identifier; no resolver link, observed 2026-07-31T12:56:46.806841Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ebebcc82-b04e-4bc9-b2d1-8c3f2fd00856 · outbound

This paper cites Distilling the Knowledge in a Neural Network.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Distilling the Knowledge in a Neural Network

Reference 2015

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.764763Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5129d7ef-986f-49e1-9161-c36d44b6fc21 · outbound

This paper cites Recursively Summarizing Enables Long-Term Dialogue Memory in Large Language Models.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Recursively Summarizing Enables Long-Term Dialogue Memory in Large Language Models

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.074813Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7fcdedf1-afc9-4a53-8d29-2ed1fc60cea6 · outbound

This paper cites Compressive Transformers for Long-Range Sequence Modelling.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Compressive Transformers for Long-Range Sequence Modelling

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.774990Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b9b10e7-af59-439b-84b4-5bccc4e2235c · outbound

This paper cites RoFormer: Enhanced Transformer with Rotary Position Embedding.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory RoFormer: Enhanced Transformer with Rotary Position Embedding

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:45.848263Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e3440b2f-b3f3-44d2-bf0a-e6d85d53dfda · outbound

This paper cites LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.371714Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f018553-5d42-4941-a98d-0fec93bc3a9b · outbound

This paper cites PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.434623Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1591ab30-e8f3-4e69-b554-2d5d89ca6e6a · outbound

This paper cites Qwen3 Technical Report.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Qwen3 Technical Report

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:46.369683Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 910c87a8-5fa8-45c2-84b1-25cfb63079ee · outbound

This paper cites Training Transformers for KV Cache Compressibility.

Understanding Is Done Early: A Depth Division of Labor in Large Language Models and Its Use for Unbounded-Context Memory Training Transformers for KV Cache Compressibility

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-07-31T12:56:44.516747Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.