Paper Citation Record · LEDGER

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs

As of 4 August 2026, this Paper Citation Record lists 12 of 12 outbound references and 1 inbound Pith citation observation for arXiv:2602.01914.

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

pith.paper-citation-record.v1
2602.01914 v4

Coverage vector

measured 12 of 12 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-03T05:36:43.087627Z

measured 13 of 13 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-04T06:34:03.388597+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-01T12:04:25.427632Z

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

12 of 12 outbound references displayed

verified exact0
verified fuzzy0
unresolved12
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation a9090290-9c80-47ba-9ac8-c6d729d33d10 · outbound

This paper cites Let wi be a scalar weight for token i (for the first hop, wi = 1; for recursive hops, wi is determined by the previous iteration).

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Let wi be a scalar weight for token i (for the first hop, wi = 1; for recursive hops, wi is determined by the previous iteration)

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:41.802598Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e8f7366-5eeb-41fc-8d4c-cb8c83d19a4a · outbound

This paper cites Doing this for all pairs (i, j)where i∈S is computationally expensive.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Doing this for all pairs (i, j)where i∈S is computationally expensive

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:41.974189Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2fdc5836-ae83-4112-81d1-6146bde915c3 · outbound

This paper cites For each layer, we calculate: •Token Importance (e tok,j):The proximity of sourcej’s contribution to the target span.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs For each layer, we calculate: •Token Importance (e tok,j):The proximity of sourcej’s contribution to the target span

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:42.141962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 88fdac9b-de5c-40fd-939f-74b4a1f019ea · outbound

This paper cites The score for token j at layer l is normalized by the sum of all token contributions plus the residual importance at that layer.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs The score for token j at layer l is normalized by the sum of all token contributions plus the residual importance at that layer

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:42.259904Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14ecd4a2-fb4b-47a1-805c-0345e344d92d · outbound

This paper cites This produces a new distribution w(k) describing which inputs caused the specific reasoning tokens that were important in the previous step.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs This produces a new distribution w(k) describing which inputs caused the specific reasoning tokens that were important in the previous step

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:42.435614Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 037b00ca-592f-4b05-89c5-5bc8858cf0cc · outbound

This paper cites an unresolved cited work.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Unresolved cited work

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:42.614194Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf23fe33-dd0f-4729-bfac-db3197f6bceb · outbound

This paper cites an unresolved cited work.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Unresolved cited work

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:42.781566Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44ec5eff-1e0f-473c-8045-fee573231127 · outbound

This paper cites an unresolved cited work.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Unresolved cited work

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:42.968507Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6c8ebbdd-f6ba-437d-aaab-48799523926a · outbound

This paper cites needles” (key-value pairs in the format “The special magic number for X is Y.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs needles” (key-value pairs in the format “The special magic number for X is Y

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:43.087627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d2593036-9456-4473-a4e0-c31132639ec8 · outbound

This paper cites Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:41.475514Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a8f48ec-76b9-43f0-9808-fc08d81c9ef7 · outbound

This paper cites Qwen3 Technical Report.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Qwen3 Technical Report

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:41.643779Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 966e6e16-3ee4-48dd-9ddd-9febaef97d0f · outbound

This paper cites Chen, J., Li, X., Yu, L., Dou, D., and Xiong, H.

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs Chen, J., Li, X., Yu, L., Dou, D., and Xiong, H

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-03T05:36:41.395452Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 7491ebde-a102-4d7e-bbbf-65bb717967b5 · inbound

GhostPrompt: Cross-Image Adversarial Prompt for Vision-Language Models cites this paper.

GhostPrompt: Cross-Image Adversarial Prompt for Vision-Language Models Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-01T12:04:25.427632Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.