Pith. sign in

Paper Citation Record · LEDGER

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models

As of 7 August 2026, this Paper Citation Record lists 31 of 31 outbound references and 1 inbound Pith citation observation for arXiv:2508.12803.

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

pith.paper-citation-record.v1
2508.12803 v1

Coverage vector

measured 31 of 31 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-05T19:23:53.473531Z

measured 32 of 32 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 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-11T22:53:25.149333Z

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

31 of 31 outbound references displayed

  • verified exact4
  • verified fuzzy3
  • unresolved24
  • parse uncertain0
  • malformed identifier0
  • metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 535ff4e8-8870-4965-8fa8-98150ed0c9f7 · outbound

This paper cites an unresolved cited work.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Unresolved cited work

Reference 1

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:49.888951Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:49.888951Z digest=sha256:5dda41bd725481b4f10d6cdb10030c2ba8f19acf0f834bf0d5b1ce5858599bfd

Observation e4554a06-3596-48ed-8737-d4cf6e9d91a5 · outbound

This paper cites Aya 23: Open Weight Releases to Further Multilingual Progress.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Aya 23: Open Weight Releases to Further Multilingual Progress

Reference 2

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:49.982686Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:49.982686Z digest=sha256:b53f344a76dc02f43539bb2507b37f504cf081ab6c4c7fdb94942270fceb31e5

Observation 41d8dcbe-8dca-4d14-8a7c-32b402608436 · outbound

This paper cites The MADAR A rabic dialect corpus and lexicon.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models The MADAR A rabic dialect corpus and lexicon

Reference 3

Resolution
verified fuzzy
raw_fallback, observed 2026-08-05T19:23:56.488958Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:50.097574Z digest=sha256:42e42fd95a0ac11f3223e5a969b67b1ec0830c42d53a85518ea356a538bec4df

Observation 98f86c66-100a-48e6-85c3-eb2f5424532f · outbound

This paper cites Language models are few-shot learners.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Language models are few-shot learners

Reference 4

Resolution
verified fuzzy
raw_fallback, observed 2026-08-05T19:23:56.211479Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:50.231222Z digest=sha256:a6d784cb0656be824ef7b3fd5cbcf70f42533c5fceb83f1a814219544eef6d0f

Observation a2144bc4-c463-48c8-a0fb-73705845a9b2 · outbound

This paper cites The geometry of multilingual language model representations.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models The geometry of multilingual language model representations

Reference 5

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:50.288724Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:50.288724Z digest=sha256:bfb3a0619a536d1626c94600f11ef74689343813ed4b06619e131b7fa545c75e

Observation 3ed83406-60c1-45c2-8930-480968724427 · outbound

This paper cites PaLM: Scaling Language Modeling with Pathways.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models PaLM: Scaling Language Modeling with Pathways

Reference 6

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:50.398604Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:50.398604Z digest=sha256:50fc0cea5137999676be3f155c50fe8f2976ca9cfa54c7358993eaa71f96cd2c

Observation c9b144b2-3a95-4e2f-80d7-59926de2b2ab · outbound

This paper cites Unsupervised Cross-lingual Representation Learning at Scale.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Unsupervised Cross-lingual Representation Learning at Scale

Reference 7

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:50.511727Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:50.511727Z digest=sha256:ef92dd4527c9fe8b866ada035db91b6e5b3a6f898c98b8b99cff224101affcc3

Observation bf161eb8-58a2-4005-b1de-003679e2ef53 · outbound

This paper cites Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier

Reference 8

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:50.609778Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:50.609778Z digest=sha256:eb560b3418a9b03617b3942135cfd36652d7f73da301fe9439a1add24360a038

Observation d65328bd-02aa-4b54-a32c-7e969f9a890a · outbound

This paper cites XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization

Reference 9

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:50.740968Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:50.740968Z digest=sha256:0b35ebab2a451515f79e947cb380ce17868131be58e430b45f9ee963ba4d1a10

Observation 1ab19e61-9d64-4510-ae4f-393bed3c4d06 · outbound

This paper cites TARJAMAT : Evaluation of bard and C hat GPT on machine translation of ten A rabic varieties.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models TARJAMAT : Evaluation of bard and C hat GPT on machine translation of ten A rabic varieties

Reference 10

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:50.897432Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:50.897432Z digest=sha256:63bcd6b5c5f9870d75be0878c022bffc7e01c87546f9e278006b2ca23a7cbfaf

Observation 5cebb704-45d3-4c1a-acf0-48fb654a799a · outbound

This paper cites Quantifying the dialect gap and its correlates across languages.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Quantifying the dialect gap and its correlates across languages

Reference 11

Resolution
verified exact
doi, observed 2026-08-05T19:23:54.376470Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:50.992661Z digest=sha256:8fcfc445f13aa21b404b195251f2663def25fd95bd370804a76759a533372f51

Observation 12e4cc46-b7f8-498f-8bae-49fec8041345 · outbound

This paper cites On the multilingual ability of decoder-based pre-trained language models: Finding and controlling language-specific neurons.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models On the multilingual ability of decoder-based pre-trained language models: Finding and controlling language-specific neurons

Reference 12

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:51.094851Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:51.094851Z digest=sha256:9eb5a247bd6fec0c3cb5767f2b67b8882896d4332018735bde6c40e510d38049

Observation aaa301ce-61bb-4698-a6fd-a6f259d1afb1 · outbound

This paper cites Subspace chronicles: How linguistic information emerges, shifts and interacts during language model training.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Subspace chronicles: How linguistic information emerges, shifts and interacts during language model training

Reference 13

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:51.174835Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:51.174835Z digest=sha256:cb205c5772ae45eee78812aa9033d1b3d66fb73155776cc0eed02e6941e3a068

Observation 38ff5a52-8d73-4f3c-835d-5b645590d1de · outbound

This paper cites Dolphin: A challenging and diverse benchmark for A rabic NLG.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Dolphin: A challenging and diverse benchmark for A rabic NLG

Reference 14

Resolution
verified exact
doi, observed 2026-08-05T19:23:54.083496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:51.304558Z digest=sha256:8088306be4929f20679aa8d9d9a06a923acb036bec2fed4d7aaf09f2f3da2236

Observation 55e288e1-f9a8-4ada-823d-41f401c571fe · outbound

This paper cites The less the merrier? investigating language representation in multilingual models.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models The less the merrier? investigating language representation in multilingual models

Reference 15

Resolution
verified exact
doi, observed 2026-08-05T19:23:53.804856Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:51.451664Z digest=sha256:1ed0f858ee0d9222b553ce18fd9f5b7dd9bc236eb45a97b75753fdaabd171ad3

Observation d7f91863-d67d-4c6d-943a-725bdb72fa4e · outbound

This paper cites chr F : character n-gram F -score for automatic MT evaluation.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models chr F : character n-gram F -score for automatic MT evaluation

Reference 16

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:51.575305Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:51.575305Z digest=sha256:60d821f59f77a80b1b3e74e7f70b9d30c0fa87cd0e7e3807612882a836aef14d

Observation 137b9974-ec29-49b1-b4af-126985e930ad · outbound

This paper cites A ra B ench: Benchmarking dialectal A rabic- E nglish machine translation.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models A ra B ench: Benchmarking dialectal A rabic- E nglish machine translation

Reference 17

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:51.701060Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:51.701060Z digest=sha256:2eab3ec498ec28f254e8340b88e9895cd3bafc3ff40e0015f69241204d3eed02

Observation 8822ac63-6ed5-4e7f-ab66-66f9f8b7b35a · outbound

This paper cites BLOOM: A 176B-Parameter Open-Access Multilingual Language Model.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models BLOOM: A 176B-Parameter Open-Access Multilingual Language Model

Reference 18

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:51.814392Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:51.814392Z digest=sha256:1be3514b060fb28d088726bd8f986c490e81468ea171a2259f1260d4e93b4e58

Observation d83811e2-f42d-46e8-8034-b7f676ad712c · outbound

This paper cites Jais and jais-chat: Arabic-centric foundation and instruction-tuned open generative large language models, 2023.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Jais and jais-chat: Arabic-centric foundation and instruction-tuned open generative large language models, 2023

Reference 19

Resolution
verified fuzzy
raw_fallback, observed 2026-08-05T19:23:55.888519Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:51.934209Z digest=sha256:0c890187b34c7d658dd90d97ce587a224c1e3ece3c746f01d436295e35397bef

Observation 9d57de8b-c18d-442c-9a7d-099b3c486405 · outbound

This paper cites an unresolved cited work.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Unresolved cited work

Reference 20

Resolution
unresolved
raw_fallback, observed 2026-08-05T19:23:55.617987Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:52.027449Z digest=sha256:bd6b7490ae3d19474c209e46bf31d993def1e789e25d4c9a802fd1226e36d5bf

Observation b95afc17-f649-478c-9340-57fd437fee67 · outbound

This paper cites an unresolved cited work.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Unresolved cited work

Reference 21

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.159309Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.159309Z digest=sha256:274f9dc1ec3a80aa27e70b475a0c5a5c49b891f685f159dab55fa0fa859c218d

Observation 4022e041-0f75-48a4-b1e0-3947423e6e05 · outbound

This paper cites Qwen3 Technical Report.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Qwen3 Technical Report

Reference 22

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.251973Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.251973Z digest=sha256:3a2b8cff8088e728c0202322e8cdbab4a09bc48f5640d7adf57b29df4e2d09c4

Observation 8df71e0c-5565-474f-9945-fc224fc0fb4e · outbound

This paper cites Information-theoretic probing with minimum description length.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Information-theoretic probing with minimum description length

Reference 23

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.372439Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.372439Z digest=sha256:7217aca29565927c5ccc8976c641fa200297a86f31a8e8ae26e72c8b0918b1f3

Observation 56a334b6-8747-4959-a96b-b397657102ed · outbound

This paper cites Sharing matters: Analysing neurons across languages and tasks in llms, 2024.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Sharing matters: Analysing neurons across languages and tasks in llms, 2024

Reference 24

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.497724Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.497724Z digest=sha256:72d1af86e879286de16309775c11b799b52735585b7012a4479471d5b1a87b20

Observation a25bceea-685a-4f8d-ae9e-b13399b30674 · outbound

This paper cites Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation

Reference 25

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.617706Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.617706Z digest=sha256:43dcdc9ed04c4bb91d6b01502fa8e4cf3a425e00f06fdbabd9146d1fd297c7a8

Observation ec51624f-086e-4063-8e35-713463d05f38 · outbound

This paper cites m T 5: A massively multilingual pre-trained text-to-text transformer.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models m T 5: A massively multilingual pre-trained text-to-text transformer

Reference 26

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.779105Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.779105Z digest=sha256:f37abc85d66c71e3c229b52b5449a1d2fb6c663aab91c9e51315b61dda166bb9

Observation bf3330a2-e975-401e-8fb9-ef2e3224d72a · outbound

This paper cites Multi- VALUE : A framework for cross-dialectal E nglish NLP.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Multi- VALUE : A framework for cross-dialectal E nglish NLP

Reference 27

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:52.949481Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:52.949481Z digest=sha256:7dd9d7022ce9d04ddc1a4182db618baeb50eaca447322090c1f2e3806f59d858

Observation 2e85aa17-3926-4505-a43e-3c182060088d · outbound

This paper cites write newline.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models write newline

Reference 28

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:53.087769Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:53.087769Z digest=sha256:27b0bc08bbecde1a04c4bdc26a5224b0eacb2c086c5d7d7c291daed4142d09cd

Observation 86f0f5ac-f0b1-4bbd-90d7-44cf00765a76 · outbound

This paper cites @esa (Ref.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models @esa (Ref

Reference 29

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:53.242610Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:53.242610Z digest=sha256:5a1abb8efdca56a36e5d5ee2c1dbf0ea3714f25e24bbbba4a0eb395335c55d08

Observation b5688a97-f610-4b41-b33c-ac13576cdcb8 · outbound

This paper cites an unresolved cited work.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models Unresolved cited work

Reference 30

Resolution
unresolved
no resolver link, observed 2026-08-05T19:23:53.355186Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-08-05T19:23:53.355186Z digest=sha256:8365d0d94670e36c3c6346d2dca5816a608de20ac79869adc5b785452d386099

Observation 32a468b7-62bc-486a-a0e5-88207130e16c · outbound

This paper cites ܆x!n0ELT]o wC э-T6S'75W ?pwKnz PML7W >oߟ ܎ -֥zZZQ=p,3Pg8O`k˄+nqq//I y H3 +ƢP Ts*.

When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models ܆x!n0ELT]o wC э-T6S'75W ?pwKnz PML7W >oߟ ܎ -֥zZZQ=p,3Pg8O`k˄+nqq//I y H3 +ƢP Ts*

Reference 31

Resolution
verified exact
raw_fallback, observed 2026-08-05T19:23:54.905655Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.

source=arxiv_source observed=2026-08-05T19:23:53.473531Z digest=sha256:769356b213a8c9c93a00cb3994ce74a7e318fda77782a2fc2cf3cc4a6a614729

Pith citing papers

Observation ec740650-302f-4117-8534-ef4ce1e2215f · inbound

Can Dialects Be Steered Like Languages? Sparse Neurons and Distributed Directions in Arabic LLMs cites this paper.

Can Dialects Be Steered Like Languages? Sparse Neurons and Distributed Directions in Arabic LLMs When Alignment Hurts: Decoupling Representational Spaces in Multilingual Models

Reference 48

Resolution
unresolved
no resolver link, observed 2026-07-11T22:53:25.149333Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=arxiv_source observed=2026-07-11T22:53:25.149333Z digest=sha256:ad8cd0502ca8f40f7c2d74b229630ebd4906772595de175ee9307ddd8f727d92