Paper Citation Record · LEDGER

Stochasticity in Tokenisation Improves Robustness

As of 4 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 1 inbound Pith citation observation for arXiv:2604.16037.

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

pith.paper-citation-record.v1
2604.16037 v1

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-10T08:49:55.126650Z

measured 25 of 25 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 2026-07-30T19:48:50.885302Z

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

24 of 24 outbound references displayed

verified exact5
verified fuzzy11
unresolved3
parse uncertain0
malformed identifier1
metadata mismatch4

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 2e873ac8-d1b5-4dd0-8977-5d11d915cbc3 · outbound

This paper cites GPT-4 Technical Report.

Stochasticity in Tokenisation Improves Robustness GPT-4 Technical Report

Reference 1

Resolution: metadata mismatch; local_arxiv, observed 2026-05-10T08:53:03.784716Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d36a589d-624c-4ae4-b494-a8c014eac9e8 · outbound

This paper cites TokSuite: Measuring the Impact of Tokenizer Choice on Language Model Behavior.

Stochasticity in Tokenisation Improves Robustness TokSuite: Measuring the Impact of Tokenizer Choice on Language Model Behavior

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-07-07T03:18:51.884510Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c4be3e40-7494-4e0b-b3ad-68286349b451 · outbound

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

Stochasticity in Tokenisation Improves Robustness Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-05-10T08:53:03.798138Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 76028606-14fc-4210-987d-f8e51d8ab2e9 · outbound

This paper cites Distributional properties of subword regularization.

Stochasticity in Tokenisation Improves Robustness Distributional properties of subword regularization

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.658322Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation f76ea9d1-9a9a-4d24-8050-877b3dbe9731 · outbound

This paper cites Attacks, defenses and evaluations for LLM conversation safety: A survey.

Stochasticity in Tokenisation Improves Robustness Attacks, defenses and evaluations for LLM conversation safety: A survey

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.682607Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c7b6e8b3-8c80-4498-aada-0b61ae235663 · outbound

This paper cites CUTE: Measuring LLMs’ understanding of their tokens.

Stochasticity in Tokenisation Improves Robustness CUTE: Measuring LLMs’ understanding of their tokens

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.660455Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation abf06e97-18ff-464a-9505-e731040e670a · outbound

This paper cites C., and Bau, D.

Stochasticity in Tokenisation Improves Robustness C., and Bau, D

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.665578Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 62df1c29-65af-4c27-95d4-11c723ed1a20 · outbound

This paper cites Where is the signal in tokenization space? InProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp.

Stochasticity in Tokenisation Improves Robustness Where is the signal in tokenization space? InProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.684769Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 50c7e07e-2dd0-4fde-b1d0-56e01f098c1a · outbound

This paper cites The Llama 3 Herd of Models.

Stochasticity in Tokenisation Improves Robustness The Llama 3 Herd of Models

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-05-10T08:53:03.794257Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 953d86d0-a663-402c-b8e0-9d5b1fc598b4 · outbound

This paper cites Super Tiny Language Models.

Stochasticity in Tokenisation Improves Robustness Super Tiny Language Models

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-10T08:53:03.778013Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 54aa63ef-c5d2-4158-a03c-314999feb6f7 · outbound

This paper cites Spelling-out is not Straightforward: LLMs' Capability of Tokenization from Token to Characters.

Stochasticity in Tokenisation Improves Robustness Spelling-out is not Straightforward: LLMs' Capability of Tokenization from Token to Characters

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-05-10T08:53:03.781857Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation b6321479-6440-44af-9a76-f052317806d6 · outbound

This paper cites and Levy, O.

Stochasticity in Tokenisation Improves Robustness and Levy, O

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.662139Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 040dd68e-b9bb-4eb5-849b-3351e4756c7c · outbound

This paper cites and Mahowald, K.

Stochasticity in Tokenisation Improves Robustness and Mahowald, K

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.680760Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 98029a11-06ba-4844-bcdd-36a32048f049 · outbound

This paper cites Red teaming language models with language models.

Stochasticity in Tokenisation Improves Robustness Red teaming language models with language models

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.663923Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c9a049a3-055c-4f6a-97f1-a553849e32d4 · outbound

This paper cites Com- monsenseQA: A question answering challenge targeting commonsense knowledge.

Stochasticity in Tokenisation Improves Robustness Com- monsenseQA: A question answering challenge targeting commonsense knowledge

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.676989Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 96d38660-6743-445a-b183-36d4ea0651d2 · outbound

This paper cites Gemini: A Family of Highly Capable Multimodal Models.

Stochasticity in Tokenisation Improves Robustness Gemini: A Family of Highly Capable Multimodal Models

Reference 16

Resolution: metadata mismatch; local_arxiv, observed 2026-05-10T08:53:03.791055Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation aac53cb6-951b-481e-a938-0e8f13264e07 · outbound

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

Stochasticity in Tokenisation Improves Robustness Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation

Reference 17

Resolution: metadata mismatch; arxiv_id, observed 2026-05-12T15:21:29.028955Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 4cca732d-2ef0-421d-aac6-90269d7129a8 · outbound

This paper cites Zheng, B.

Stochasticity in Tokenisation Improves Robustness Zheng, B

Reference 18

Resolution: metadata mismatch; arxiv_id, observed 2026-05-10T08:53:03.802001Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ea20393d-d236-44d3-b11c-ecb6eff42b6f · outbound

This paper cites Training Details /githubThe code for the experiments is available at: https://github.com/stegsoph/ stochastic-tokenisation-robustness.

Stochasticity in Tokenisation Improves Robustness Training Details /githubThe code for the experiments is available at: https://github.com/stegsoph/ stochastic-tokenisation-robustness

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.670951Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d10d72c4-7513-4ed6-b4cd-2976fa68ef9a · outbound

This paper cites an unresolved cited work.

Stochasticity in Tokenisation Improves Robustness Unresolved cited work

Reference 20

Resolution: unresolved; raw_fallback, observed 2026-05-20T19:23:55.672678Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 6753f6d9-2780-4b78-92ea-6fe4ea5797b1 · outbound

This paper cites - stood-.

Stochasticity in Tokenisation Improves Robustness - stood-

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T19:23:55.674401Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 4a1ca95e-2ec8-4b66-8fe6-e86f337aa55b · outbound

This paper cites an unresolved cited work.

Stochasticity in Tokenisation Improves Robustness Unresolved cited work

Reference 22

Resolution: unresolved; raw_fallback, observed 2026-05-20T19:23:55.678809Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 6860f2ff-66f4-42d6-b9f0-04e70a9b1d38 · outbound

This paper cites an unresolved cited work.

Stochasticity in Tokenisation Improves Robustness Unresolved cited work

Reference 23

Resolution: unresolved; raw_fallback, observed 2026-05-20T19:23:55.687087Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation cf3b7f56-3c74-4617-9daa-ec1b7a4ab603 · outbound

This paper cites Importantly, during stochastic pre-training, the model is trained to predict non-canonical token sequences.

Stochasticity in Tokenisation Improves Robustness Importantly, during stochastic pre-training, the model is trained to predict non-canonical token sequences

Reference 24

Resolution: malformed identifier; raw_fallback, observed 2026-05-20T19:23:55.668162Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Pith citing papers

Observation 5f4abd28-be4b-41d4-bc95-daffea586e02 · inbound

Language Models are not Equally Robust to Non-Canonical Tokenization across Languages cites this paper.

Language Models are not Equally Robust to Non-Canonical Tokenization across Languages Stochasticity in Tokenisation Improves Robustness

Reference 63

Resolution: unresolved; no resolver link, observed 2026-07-30T19:48:50.885302Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.