Pith. sign in

Paper Citation Record · LEDGER

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs

As of 9 August 2026, this Paper Citation Record lists 27 of 27 outbound references and 0 inbound Pith citation observations for arXiv:2506.08543.

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

pith.paper-citation-record.v1
2506.08543 v4

Coverage vector

measured 27 of 27 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T05:18:32.081506Z

measured 27 of 27 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+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

27 of 27 outbound references displayed

  • verified exact1
  • verified fuzzy5
  • unresolved21
  • parse uncertain0
  • malformed identifier0
  • metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 9ca964d1-3ae0-4e92-848c-64dc6ccd3df7 · outbound

This paper cites Then, with probability at least1−2 exp(−t 2/2), η(l) ≤ τ2 d + B2 (t+ √ d)√ M·d.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Then, with probability at least1−2 exp(−t 2/2), η(l) ≤ τ2 d + B2 (t+ √ d)√ M·d

Reference 2

Resolution
verified fuzzy
raw_fallback, observed 2026-08-07T05:18:33.033582Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:31.936392Z digest=sha256:24b05cb4c029921e93e58565208a2b9cb2bdc6894f8fa1df0335872de4736e82

Observation 6cc488f7-87cb-44a7-b6e1-af6a9e6bcf7e · outbound

This paper cites Evaluating Large Language Models Trained on Code.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Evaluating Large Language Models Trained on Code

Reference 3

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:29.081026Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:29.081026Z digest=sha256:35003c01255e8ce4b80153980e5d55bf24e1ae219aa7202d49e0fd1fd55d5b0c

Observation 6eb2c334-df9c-4102-a0ac-c1189336dc19 · outbound

This paper cites 19 Under review as a conference paper at COLM 2026 Figure 6: VLM response for power concept with U.S.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs 19 Under review as a conference paper at COLM 2026 Figure 6: VLM response for power concept with U.S

Reference 4

Resolution
verified fuzzy
raw_fallback, observed 2026-08-07T05:18:32.739887Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:32.081506Z digest=sha256:fddaa8d17872a7b516a494949f82ce06d5faa855c070fb3bce3f48e58dbd6520

Observation ae477b81-fcb5-4a16-9393-722e49937a88 · outbound

This paper cites The Llama 3 Herd of Models.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs The Llama 3 Herd of Models

Reference 6

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:29.459977Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:29.459977Z digest=sha256:5268e0b6d000fa5b18724ef4778857555f6ec12a9b0259dfe11ed30c3644f5f4

Observation b6bab8a0-4473-4f04-9df1-0185099711ed · outbound

This paper cites an unresolved cited work.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Unresolved cited work

Reference 9

Resolution
unresolved
raw_fallback, observed 2026-08-07T05:18:34.037673Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:29.930887Z digest=sha256:a22b3b36405c2938987e83d9cff3fb430d46dd2931136bb149556268b347dc0e

Observation eae13975-fe78-4afc-ba06-01077306ed50 · outbound

This paper cites Mistral 7B.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Mistral 7B

Reference 10

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.012097Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.012097Z digest=sha256:b5a4d9ca9d93b78b6ab3e39ace3f652e698ba026d142dff82fd7224ca24a4524

Observation f2998c2a-9769-4de5-b683-b5abb38ca70f · outbound

This paper cites Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla

Reference 11

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.103621Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.103621Z digest=sha256:38ed5cbed9d5fcee8365669b98e39808790616739103dd20fb83bc9f7d359d6d

Observation 6377fa61-6165-4e4d-8dcf-de6a3138a832 · outbound

This paper cites Microsoft coco: Common objects in context.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Microsoft coco: Common objects in context

Reference 13

Resolution
verified fuzzy
raw_fallback, observed 2026-08-07T05:18:33.636921Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:30.372307Z digest=sha256:d3454be8bff06849549e87c27deda4a9932fa15252e9dcc81e5481e96a77f41c

Observation ec61e66d-925e-4c1d-a304-36fb0c470ebc · outbound

This paper cites DINOv2: Learning Robust Visual Features without Supervision.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs DINOv2: Learning Robust Visual Features without Supervision

Reference 16

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.777638Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.777638Z digest=sha256:f591856397da003b4606aa5f57a3113f8c33440fa08f00ddc2bb449470b76b24

Observation a06498c6-173e-4269-b152-c2b5b8a40e74 · outbound

This paper cites The Linear Representation Hypothesis and the Geometry of Large Language Models.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs The Linear Representation Hypothesis and the Geometry of Large Language Models

Reference 17

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.862403Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.862403Z digest=sha256:ff4b3dde791b0dce62a2ca5cff3c1c7d3388b5682d73f62f2c469288ede5798d

Observation 26123c9b-7a44-46f8-88f7-99b708be8a90 · outbound

This paper cites Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

Reference 18

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.952553Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.952553Z digest=sha256:6943925ed3f4b0dae2055a54dba09e0ce3186aa754351e296780084e3eef414a

Observation 75fe3fbc-062e-4e53-a1e3-bb30fefa02d7 · outbound

This paper cites Striving for Simplicity: The All Convolutional Net.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Striving for Simplicity: The All Convolutional Net

Reference 21

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:31.327333Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:31.327333Z digest=sha256:bc3e9006adc9498c1edb53c95031fc2e02528edcb3bd3101cc283cc9189e6fa8

Observation 299a496d-dd30-44b1-b947-2b722fc5eaea · outbound

This paper cites Ethical and social risks of harm from Language Models.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Ethical and social risks of harm from Language Models

Reference 23

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:31.568017Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:31.568017Z digest=sha256:8e0342baf6cbd661d7844e73311a05b212a3e25b6139c874a7d94e3c82b8173f

Observation 3f7b832d-b14b-41c2-b820-2a0da246a658 · outbound

This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Representation Engineering: A Top-Down Approach to AI Transparency

Reference 24

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:31.670633Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:31.670633Z digest=sha256:67c91f482504dc4a48c5e28f83b43635aaa603e51fb5b1ef37f8e9ddc3035816

Observation d9248ace-109b-4eef-9f01-2065927d44ce · outbound

This paper cites signal” matrix D(l) ¯J(l)⊤ and a “perturbation.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs signal” matrix D(l) ¯J(l)⊤ and a “perturbation

Reference 25

Resolution
verified fuzzy
raw_fallback, observed 2026-08-07T05:18:33.386347Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:31.787696Z digest=sha256:01a57f0d6c39a0a5543c15c5082aecbac661216fd25a8ae768aab53702cf42be

Observation e41a07be-e189-45f1-bc37-d576af203959 · outbound

This paper cites SimCSE: Simple Contrastive Learning of Sentence Embeddings.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs SimCSE: Simple Contrastive Learning of Sentence Embeddings

Reference 1959

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:29.374638Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:29.374638Z digest=sha256:1e9bc11b80bdafadd6c153e8e9982c6ef726239df96734d3b3c7daf0099fab5d

Observation f07afe15-3513-4348-a9d7-d030287b3901 · outbound

This paper cites TruthfulQA: Measuring How Models Mimic Human Falsehoods.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs TruthfulQA: Measuring How Models Mimic Human Falsehoods

Reference 1996

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.243714Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.243714Z digest=sha256:749f2a64a87ad0df5c11f91d95fa9e0c63f33061f1cff6916cbbdba78ddbb9e6

Observation 5b5e8603-719c-46ed-8431-955248007978 · outbound

This paper cites Progress measures for grokking via mechanistic interpretability.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Progress measures for grokking via mechanistic interpretability

Reference 2013

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.500734Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.500734Z digest=sha256:cd01afce66ef31e8edb63d9c3e4c0a612ddf8731c062ff55cacb0e15ee189b03

Observation b7ed31b8-7b1c-4cf3-941e-147189b51063 · outbound

This paper cites Intriguing properties of neural networks.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Intriguing properties of neural networks

Reference 2014

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:31.407979Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:31.407979Z digest=sha256:32028361f7d4b160edd3e2ce649dcfdcb8b2c3a60eb80a1290293c9913d607f1

Observation 9af1b8e1-e234-4ac1-adad-c4786f2216b8 · outbound

This paper cites Qwen2.5-Coder Technical Report.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Qwen2.5-Coder Technical Report

Reference 2016

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:29.790144Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:29.790144Z digest=sha256:25875fe1442603b215b9820bc62ff383ad2f6317dc083f8175fdc6b6b08253f6

Observation 452cb83f-6498-4bb3-98e5-279e3ac8d63b · outbound

This paper cites BERT has a Moral Compass: Improvements of ethical and moral values of machines.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs BERT has a Moral Compass: Improvements of ethical and moral values of machines

Reference 2017

Resolution
verified exact
local_arxiv, observed 2026-08-07T05:18:32.412798Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:31.068471Z digest=sha256:71c4ca062231fc42268593cfc3c289dee5dcf299e281aac70e2c142eebc7097b

Observation c8251af8-2a95-4a99-ae5a-83905f462f54 · outbound

This paper cites Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

Reference 2019

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:31.180269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:31.180269Z digest=sha256:3d63e25c696cc88ad89962f7e7174034a3f702c235ba355cc6a8da63abd1a454

Observation 77c2a1aa-fbef-4047-bbad-a4df96f0970b · outbound

This paper cites In-context Learning and Induction Heads.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs In-context Learning and Induction Heads

Reference 2020

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:30.658144Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:30.658144Z digest=sha256:b203dd960561836e5abbc76258cac0d1051fc825dc5dde4342cc34c952b126ef

Observation 35c77375-8a76-4304-8db5-4449fec52c29 · outbound

This paper cites Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901,.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Language models are few-shot learners.Advances in neural information processing systems, 33:1877–1901,

Reference 2021

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:29.002343Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:29.002343Z digest=sha256:b69b6002cef1648d7e96ad10190b0f636550aff278e13e0fec10d20974a3b5c6

Observation b43e6f72-05ba-4536-99c3-14f91f06be9d · outbound

This paper cites On the Opportunities and Risks of Foundation Models.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs On the Opportunities and Risks of Foundation Models

Reference 2022

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:28.898579Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:28.898579Z digest=sha256:7cbfc921a23c49a33b5620a0dd36b2c90a56b9fbf016a460ad8c1460c835b1fa

Observation 611ecb9d-ebce-417d-9571-c03aa69d3cee · outbound

This paper cites Bert: Pre-training of deep bidirectional transformers for language understanding.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Bert: Pre-training of deep bidirectional transformers for language understanding

Reference 2023

Resolution
verified fuzzy
raw_fallback, observed 2026-08-07T05:18:34.150724Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

source=pdf_text observed=2026-08-07T05:18:29.208342Z digest=sha256:27041752edf272c493f1203a530bdf64e849f92b43db3f3cc38dc4454f761214

Observation ec3b3428-190f-4ede-8df4-b727a834e3d2 · outbound

This paper cites Language Models Represent Space and Time.

Spectral Principal Paths: A Spectral Perspective on Linear Representation Formation in LLMs Language Models Represent Space and Time

Reference 2024

Resolution
unresolved
no resolver link, observed 2026-08-07T05:18:29.599765Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-07T05:18:29.599765Z digest=sha256:c1509d745f564e8876b1411670f09ad8941c6e6ff95aae94a3ba919fc754c110

Pith citing papers

No inbound Pith citation observations are available.