Paper Citation Record · LEDGER

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

As of 20 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 5 inbound Pith citation observations for arXiv:2507.03865.

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

pith.paper-citation-record.v1
2507.03865 v2

Coverage vector

measured 40 of 40 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-06T20:08:09.488803Z

measured 45 of 45 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-20T06:33:59.587034+00:00

measured 5 of 5 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-05T13:15:07.167572Z

measured 1 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Reference resolution

40 of 40 outbound references displayed

verified exact0
verified fuzzy18
unresolved22
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

0
arxiv_reference, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation 8025c31f-e790-499d-bb46-40311bd8d76e · outbound

This paper cites write newline.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference write newline

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.520040Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3271328d-b004-4830-869d-45401ba497cd · outbound

This paper cites Fast and robust early-exiting framework for autoregressive language models with synchronized parallel decoding.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Fast and robust early-exiting framework for autoregressive language models with synchronized parallel decoding

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.144777Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 4861d466-5830-4fd9-b861-0622edabd0d0 · outbound

This paper cites Longbench: A bilingual, multitask benchmark for long context understanding.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Longbench: A bilingual, multitask benchmark for long context understanding

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.134965Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation a3febf6d-418c-4022-a485-f1c4592d3c68 · outbound

This paper cites Piqa: Reasoning about physical commonsense in natural language.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Piqa: Reasoning about physical commonsense in natural language

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.665742Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b46542b-851f-463d-8eef-d25b67658055 · outbound

This paper cites Spectral filters, dark signals, and attention sinks.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Spectral filters, dark signals, and attention sinks

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.694850Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7e3cbe33-ce91-4136-8db3-9c084ff0c855 · outbound

This paper cites SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.728594Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3871d836-3f1b-46ac-99f7-e2d643164297 · outbound

This paper cites Ee-llm: Large-scale training and inference of early-exit large language models with 3d parallelism.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Ee-llm: Large-scale training and inference of early-exit large language models with 3d parallelism

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.119715Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 6a2e375d-4bd9-46bc-b2f7-ceee1f13df11 · outbound

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

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.826197Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8fe2c9d0-f948-465b-942a-3381ee8bc030 · outbound

This paper cites SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.884228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22d001fd-06b5-44b3-9188-35a713a1b314 · outbound

This paper cites The Llama 3 Herd of Models.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference The Llama 3 Herd of Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:06.940790Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf401c8b-e256-4a47-9367-6c7a4db9f037 · outbound

This paper cites Layerskip: Enabling early exit inference and self-speculative decoding.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Layerskip: Enabling early exit inference and self-speculative decoding

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.110224Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 1c73407b-f2ef-4a55-b6bb-09fa8cbfb7b4 · outbound

This paper cites When attention sink emerges in language models: An empirical view.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference When attention sink emerges in language models: An empirical view

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.099806Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 870b8fee-af36-484a-8381-11b1f0f59ac2 · outbound

This paper cites J., yelong shen, Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference J., yelong shen, Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.136253Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3133fe33-7106-4036-8df3-ea59d7c6b30b · outbound

This paper cites Mistral 7B.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Mistral 7B

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.195766Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 52bdaca1-fa16-4b5d-b930-9629ec47723e · outbound

This paper cites Mixtral of Experts.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Mixtral of Experts

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.249338Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c5ab7db0-7fb5-4b79-a78f-e3774e77afb1 · outbound

This paper cites D-llm: A token adaptive computing resource allocation strategy for large language models.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference D-llm: A token adaptive computing resource allocation strategy for large language models

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.082034Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation b92bd581-f141-4ed6-955b-46c974e5ccc7 · outbound

This paper cites Shortened LLaMA: Depth Pruning for Large Language Models with Comparison of Retraining Methods.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Shortened LLaMA: Depth Pruning for Large Language Models with Comparison of Retraining Methods

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.404789Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d0c73403-f92f-4756-93ab-f13daf0b7375 · outbound

This paper cites B io M istral: A collection of open-source pretrained large language models for medical domains.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference B io M istral: A collection of open-source pretrained large language models for medical domains

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.071971Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 00e02adc-9716-4445-82a2-57fffb6784c8 · outbound

This paper cites Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.633297Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ebc7bd35-3c72-41ea-8248-cfa28d1b62d4 · outbound

This paper cites ShortGPT: Layers in Large Language Models are More Redundant Than You Expect.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference ShortGPT: Layers in Large Language Models are More Redundant Than You Expect

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.685132Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5ce3c25-d6bf-44cd-8c55-7a4c02fc4a5f · outbound

This paper cites Pointer sentinel mixture models.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Pointer sentinel mixture models

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.061210Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 7e696759-321d-482e-a7b2-db9808987c1a · outbound

This paper cites Using an llm to help with code understanding.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Using an llm to help with code understanding

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.050576Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e3f9296e-de56-4039-9696-d66615f753ec · outbound

This paper cites an unresolved cited work.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Unresolved cited work

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.846759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 57f4bdd9-3066-46f7-b105-9bcddaf9047a · outbound

This paper cites Mixture-of-Depths: Dynamically allocating compute in transformer-based language models.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.957269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5224ed9-8b43-4c75-ac39-c02ce6f43d03 · outbound

This paper cites L., Bhagavatula, C., and Choi, Y.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference L., Bhagavatula, C., and Choi, Y

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:07.997272Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 67011f70-86fa-4438-b248-fe500a267bcb · outbound

This paper cites Q., Tay, Y., and Metzler, D.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Q., Tay, Y., and Metzler, D

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:11.027343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 06e44b8d-493f-4b07-9890-2f66f876fd32 · outbound

This paper cites A deeper look at depth pruning of LLMs.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference A deeper look at depth pruning of LLMs

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:08.244465Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e2bb6b72-cc2a-47c0-9b84-e450ce09f459 · outbound

This paper cites Prefixing Attention Sinks can Mitigate Activation Outliers for Large Language Model Quantization.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Prefixing Attention Sinks can Mitigate Activation Outliers for Large Language Model Quantization

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:08.298473Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 32042cfd-8cb9-4108-b8ba-e4bf8c82d96f · outbound

This paper cites Sleb: Streamlining llms through redundancy verification and elimination of transformer blocks.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Sleb: Streamlining llms through redundancy verification and elimination of transformer blocks

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.915706Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation c8970a42-2ccc-4f37-8dd1-ff8b0dd3fc6e · outbound

This paper cites Z., and Liu, Z.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Z., and Liu, Z

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.906428Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 6fc4520c-285e-49a0-b59a-02cabd591125 · outbound

This paper cites Razorattention: Efficient kv cache compression through retrieval heads.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Razorattention: Efficient kv cache compression through retrieval heads

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.896623Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 3846996a-fad5-4ef1-9520-6dc86ad22fcc · outbound

This paper cites J., Ting, D.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference J., Ting, D

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.740169Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 4533b7cf-cd02-4b68-9921-77c3bb68dc7f · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:08.722523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bd2fa558-17d1-4cef-9907-f074a86bf813 · outbound

This paper cites BloombergGPT: A Large Language Model for Finance.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference BloombergGPT: A Large Language Model for Finance

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:08.837488Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2877ea30-5aea-4eb7-8810-eddfceecbbbc · outbound

This paper cites T., Peng, R., Wu, Q., and Wang, C.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference T., Peng, R., Wu, Q., and Wang, C

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.551322Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 13964cca-0020-4437-8b2f-7895d5678333 · outbound

This paper cites Duoattention: Efficient long-context llm inference with retrieval and streaming heads.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Duoattention: Efficient long-context llm inference with retrieval and streaming heads

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.339924Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 8527fb09-f508-40c1-a700-609a7a9bc288 · outbound

This paper cites Efficient streaming language models with attention sinks.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Efficient streaming language models with attention sinks

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:10.169486Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 3ee6e21e-26f5-49f1-a821-95061880a634 · outbound

This paper cites an unresolved cited work.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Unresolved cited work

Reference 38

Resolution: unresolved; raw_fallback, observed 2026-08-06T20:08:10.029542Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 7f2a254e-64b3-4b64-a42b-e0c7054921df · outbound

This paper cites Hellaswag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp.\ 4791--4800, 2019.

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference Hellaswag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp.\ 4791--4800, 2019

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-06T20:08:09.806985Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation ae04f335-09c4-4713-9b2e-f61df35db73e · outbound

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

OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-06T20:08:09.488803Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 4040624e-1a52-449c-ad62-3e59d3d9a76d · inbound

DTRNet: Dynamic Token Routing Network to Reduce Quadratic Costs in Transformers cites this paper.

DTRNet: Dynamic Token Routing Network to Reduce Quadratic Costs in Transformers OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-05T13:15:07.167572Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1b256f7d-5bfb-4f70-a411-d8e74ae2a03f · inbound

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse cites this paper.

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

Reference 29

Resolution: verified exact; arxiv_id, observed 2026-05-16T08:47:37.280027Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e4d1dbc0-a3f4-46bd-87cd-26436f2198d6 · inbound

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse cites this paper.

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-03T05:50:25.027494Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df018183-1fe6-4e00-831b-dc59de7a9fd1 · inbound

Contribution Weights: A Geometrical Analysis of Self-Attention Transformers cites this paper.

Contribution Weights: A Geometrical Analysis of Self-Attention Transformers OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

Reference 14

Resolution: metadata mismatch; arxiv_id, observed 2026-06-28T23:32:46.707032Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e37ca751-0043-4a4b-9ad3-43f1489a797d · inbound

Contribution Weights: A Geometrical Analysis of Self-Attention Transformers cites this paper.

Contribution Weights: A Geometrical Analysis of Self-Attention Transformers OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference

Reference 171

Resolution: verified exact; arxiv_id, observed 2026-06-28T23:32:47.283100Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.