Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES | Paper Citation Record

Observation 1940799e-0e73-4639-8a0a-6917d9271468 · outbound

This paper cites Knowledge-Centric Hallucination Detection.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Knowledge-Centric Hallucination Detection

Reference 1

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.800090Z

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.

Observation fc838ed1-31b9-4fe7-9c19-7caedcaf264d · outbound

This paper cites Stop Taking Tokenizers for Granted: They Are Core Design Decisions in Large Language Models.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Stop Taking Tokenizers for Granted: They Are Core Design Decisions in Large Language Models

Reference 2

Resolution: verified exact; doi, observed 2026-07-11T03:47:47.010713Z

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.

Observation 4da1c91a-caf6-450d-975e-b003a8b086e7 · outbound

This paper cites Randomized SMILES strings improve the quality of molecular generative models.J.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Randomized SMILES strings improve the quality of molecular generative models.J

Reference 3

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.878624Z

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.

Observation e747b5e6-1036-45ef-8c62-fb459b3c9792 · outbound

This paper cites 2020 , note =.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES 2020 , note =

Reference 4

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.600036Z

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.

Observation 90503f89-3e93-45c7-85e5-157611c905a0 · outbound

This paper cites Ellie Pavlick and Tom Kwiatkowski.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Ellie Pavlick and Tom Kwiatkowski

Reference 5

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.306603Z

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.

Observation b32ebd05-36a3-4f29-a6e0-f7deb83a358a · outbound

This paper cites BARTSmiles: Generative masked language models for molecular representations.Journal of Chemical Information and Modeling, 64(15):5832– 5843, 2024.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES BARTSmiles: Generative masked language models for molecular representations.Journal of Chemical Information and Modeling, 64(15):5832– 5843, 2024

Reference 6

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.855731Z

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.

Observation 8a65a5e0-d00b-4194-b9b1-d775287052a9 · outbound

This paper cites Chemberta: Large-scale self-supervised pretraining for molecular property prediction, 2020.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Chemberta: Large-scale self-supervised pretraining for molecular property prediction, 2020

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.567331Z

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.

Observation f2743ddb-1d27-4399-9069-321606e3a725 · outbound

This paper cites Two counterexamples to tokenization and the noiseless channel.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Two counterexamples to tokenization and the noiseless channel

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.542165Z

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.

Observation c5a3a424-1c43-4216-a404-98aac6de07bf · outbound

This paper cites Investigating the effectiveness of BPE: The power of shorter sequences.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Investigating the effectiveness of BPE: The power of shorter sequences

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.605978Z

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.

Observation 87b02662-1458-40e4-9da3-7c77f1dfc0d3 · outbound

This paper cites Investigating the Effectiveness of BPE : The Power of Shorter Sequences.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Investigating the Effectiveness of BPE : The Power of Shorter Sequences

Reference 10

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.369524Z

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.

Observation 6b296931-3b02-4c47-9688-a014f4c1db5f · outbound

This paper cites Measuring chemical LLM robustness to molecular representations: a SMILES variation-based framework.Journal of Cheminformatics, 17(1):164,.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Measuring chemical LLM robustness to molecular representations: a SMILES variation-based framework.Journal of Cheminformatics, 17(1):164,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.516143Z

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.

Observation d0e1adae-2431-4eb0-a478-b2157b555c83 · outbound

This paper cites Measuring chemical LLM robustness to molecular representations: a SMILES variation-based framework.J.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Measuring chemical LLM robustness to molecular representations: a SMILES variation-based framework.J

Reference 12

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.653620Z

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.

Observation 8d8eecee-2b4f-4814-952e-b2e700fe3425 · outbound

This paper cites Finding the Optimal Vocabulary Size for Neural Machine Translation.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Finding the Optimal Vocabulary Size for Neural Machine Translation

Reference 13

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.458625Z

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.

Observation c5242c0c-59b7-4848-bf5f-45109230c5dc · outbound

This paper cites Grygorenko, Dmytro S.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Grygorenko, Dmytro S

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-07-11T03:47:46.958225Z

Source-reported events for the cited work

correction dated 2020-12-04. Source: crossref record 10.1016/j.isci.2020.101873->10.1016/j.isci.2020.101681:correction, observed 2026-07-11T03:12:21.396765+00:00. This notice travels one citation hop only.

Observation c8011620-85f0-45dc-9165-22b65608add0 · outbound

This paper cites Smirk fork for the vocabulary–tokenizer comparison study: shared glyph-id front-end with GpeTrainer (bpe) and a unigram-lm sibling trainer.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Smirk fork for the vocabulary–tokenizer comparison study: shared glyph-id front-end with GpeTrainer (bpe) and a unigram-lm sibling trainer

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.630194Z

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.

Observation d5766850-77ea-4625-ae20-1d4e9ab4e451 · outbound

This paper cites Dynamic Chunking for End-to-End Hierarchical Sequence Modeling.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Dynamic Chunking for End-to-End Hierarchical Sequence Modeling

Reference 16

Resolution: metadata mismatch; local_arxiv, observed 2026-07-11T03:47:48.606361Z

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.

Observation 2e69c206-158f-461b-ad73-256544162c11 · outbound

This paper cites an unresolved cited work.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Unresolved cited work

Reference 17

Resolution: unresolved; raw_fallback, observed 2026-07-11T03:47:52.613299Z

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.

Observation 2c8da5db-f55c-4866-8df6-cb5201df592f · outbound

This paper cites The tokenization bottleneck: How vocabulary extension improves chemistry representation learning in pretrained language models, 2025.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES The tokenization bottleneck: How vocabulary extension improves chemistry representation learning in pretrained language models, 2025

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.578748Z

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.

Observation 15670ed5-18b3-47c7-b6e3-99b77615ef55 · outbound

This paper cites Shoemaker, Paul A.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Shoemaker, Paul A

Reference 19

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.706233Z

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.

Observation 017ae20d-1532-4077-bd45-471b51371ea5 · outbound

This paper cites Self-Referencing Embedded Strings (SELFIES): A 100% Robust Molecular String Representation.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Self-Referencing Embedded Strings (SELFIES): A 100% Robust Molecular String Representation

Reference 20

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.988060Z

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.

Observation 5269996b-4cfc-4851-8f7b-638709bbe410 · outbound

This paper cites Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates

Reference 21

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.638522Z

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.

Observation 571a6f32-97b8-4183-9d0b-99818882bc21 · outbound

This paper cites SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing

Reference 22

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.758373Z

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.

Observation 40ee98c9-4b8f-451f-b0ab-8d787553f7ad · outbound

This paper cites Fishing for Magikarp: Automatically Detecting Under-trained Tokens in Large Language Models.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Fishing for Magikarp: Automatically Detecting Under-trained Tokens in Large Language Models

Reference 23

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.694437Z

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.

Observation 447ab5a8-70a8-4ab9-a7a8-bf4cf8fcd373 · outbound

This paper cites Scalfani and Yakov Pechersky and Kazuya Ujihara and Daniel Probst and Jeremy Monat and Juuso Lehtivarjo , title =.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Scalfani and Yakov Pechersky and Kazuya Ujihara and Daniel Probst and Jeremy Monat and Juuso Lehtivarjo , title =

Reference 24

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.432262Z

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.

Observation 93b3af8e-9fb0-4840-8cc4-62f930f2e312 · outbound

This paper cites Comparing SMILES and SELFIES tokenization for enhanced chemical language modeling.Scientific Reports, 14(1):25016, 2024.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Comparing SMILES and SELFIES tokenization for enhanced chemical language modeling.Scientific Reports, 14(1):25016, 2024

Reference 25

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.932430Z

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.

Observation 5d7d37ee-b699-4ab4-924a-8314de86aad2 · outbound

This paper cites CycPeptM- PDB: A comprehensive database of membrane permeability of cyclic peptides.Journal of Chemical Information and Modeling, 63(7):2240–2250, 2023.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES CycPeptM- PDB: A comprehensive database of membrane permeability of cyclic peptides.Journal of Chemical Information and Modeling, 63(7):2240–2250, 2023

Reference 26

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.732281Z

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.

Observation d8c6c11c-9266-496e-a0d1-cb6929004913 · outbound

This paper cites SMILES pair encoding: A data-driven substructure tokenization algorithm for deep learning.Journal of Chemical Information and Modeling, 61(4):1560–1569, 2021.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES SMILES pair encoding: A data-driven substructure tokenization algorithm for deep learning.Journal of Chemical Information and Modeling, 61(4):1560–1569, 2021

Reference 27

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.906150Z

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.

Observation aa1383ad-533b-4fbf-9ce6-ddf2b5e4bf44 · outbound

This paper cites Scaffold-BPE: Enhancing Byte Pair Encoding for Large Language Models with Simple and Effective Scaffold Token Removal.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Scaffold-BPE: Enhancing Byte Pair Encoding for Large Language Models with Simple and Effective Scaffold Token Removal

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-07-11T03:47:48.550408Z

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.

Observation 64990453-bd85-4645-8489-16cee6ea418f · outbound

This paper cites SuperBPE: Space Travel for Language Models.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES SuperBPE: Space Travel for Language Models

Reference 29

Resolution: metadata mismatch; local_arxiv, observed 2026-07-11T03:47:48.553716Z

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.

Observation b9b0fa1a-3e17-4e50-a9f9-477f6a31213f · outbound

This paper cites HuggingFace’s tokenizers: Fast state-of-the-art tokenizers optimized for research and production.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES HuggingFace’s tokenizers: Fast state-of-the-art tokenizers optimized for research and production

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.711742Z

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.

Observation c8705c67-22f6-4b94-8b48-f2e04376b670 · outbound

This paper cites Hierarchical Autoregressive Transformers: Combining Byte- and Word-Level Processing for Robust, Adaptable Language Models.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Hierarchical Autoregressive Transformers: Combining Byte- and Word-Level Processing for Robust, Adaptable Language Models

Reference 31

Resolution: verified exact; local_arxiv, observed 2026-07-11T03:47:48.584286Z

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.

Observation c9ddf76a-e7a6-4842-9e83-471f432030fa · outbound

This paper cites an unresolved cited work.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Unresolved cited work

Reference 32

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.772764Z

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.

Observation 445a4887-3dfb-424f-8792-47608d75d32b · outbound

This paper cites O’Boyle and Andrew Dalke.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES O’Boyle and Andrew Dalke

Reference 33

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.822759Z

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.

Observation 05d66f14-9838-4e25-9a40-c438a24c5c58 · outbound

This paper cites Byte Latent Transformer: Patches Scale Better Than Tokens.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Byte Latent Transformer: Patches Scale Better Than Tokens

Reference 34

Resolution: metadata mismatch; local_arxiv, observed 2026-07-11T03:47:48.485819Z

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.

Observation 64530f2c-2391-4eb3-9f66-65817d9f2cdd · outbound

This paper cites 2020 , journal =.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES 2020 , journal =

Reference 35

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.571580Z

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.

Observation 52cac4bd-5d78-4dd6-9751-a24a18b8ef28 · outbound

This paper cites Optimizing SMILES token sequences via trie-based refinement and transition graph filtering.Journal of Cheminformatics, 18(1):13, 2026.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Optimizing SMILES token sequences via trie-based refinement and transition graph filtering.Journal of Cheminformatics, 18(1):13, 2026

Reference 36

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.749898Z

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.

Observation b84a5871-00b0-43ff-871e-62082d5f1496 · outbound

This paper cites O’Reilly Media, 2019.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES O’Reilly Media, 2019

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.734509Z

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.

Observation dac804ef-6bcb-43dd-8deb-5fce4d35e3e0 · outbound

This paper cites How Much is Enough? The Diminishing Returns of Tokenization Training Data.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES How Much is Enough? The Diminishing Returns of Tokenization Training Data

Reference 38

Resolution: metadata mismatch; local_arxiv, observed 2026-07-11T03:47:48.502700Z

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.

Observation 439a5b4a-94d3-4980-a915-7adcaec10880 · outbound

This paper cites How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models , booktitle =.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models , booktitle =

Reference 39

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.928886Z

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.

Observation 0ad5d913-b6f6-4b78-9128-114754b17405 · outbound

This paper cites ReactionT5: a large-scale pre-trained model towards application of limited reaction data.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES ReactionT5: a large-scale pre-trained model towards application of limited reaction data

Reference 40

Resolution: verified exact; local_arxiv, observed 2026-07-11T03:47:46.825551Z

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.

Observation dbcae29c-69bb-4317-be48-03d1300d9366 · outbound

This paper cites Schmidt, Varshini Reddy, Haoran Zhang, Alec Alameddine, Omri Uzan, Yuval Pinter, and Chris C.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Schmidt, Varshini Reddy, Haoran Zhang, Alec Alameddine, Omri Uzan, Yuval Pinter, and Chris C

Reference 41

Resolution: malformed identifier; raw_fallback, observed 2026-07-11T03:47:52.757696Z

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.

Observation 991dc72e-7aef-41e6-a93e-cd8c6bb89ca7 · outbound

This paper cites Schmidt, Varshini Reddy, Chris Tanner, and Yuval Pinter.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Schmidt, Varshini Reddy, Chris Tanner, and Yuval Pinter

Reference 42

Resolution: verified exact; arxiv_id, observed 2026-07-11T03:47:48.613839Z

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.

Observation 61918727-984f-4392-8937-2015451fc2fa · outbound

This paper cites Small-footprint keyword spotting using deep neural networks.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Small-footprint keyword spotting using deep neural networks

Reference 43

Resolution: malformed identifier; doi_truncated, observed 2026-07-11T03:47:46.981057Z

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.

Observation f87d3533-8a22-43f0-86d1-549e0077e730 · outbound

This paper cites found in translation.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES found in translation

Reference 44

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.306101Z

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.

Observation 50489d37-0294-43b9-9855-51a6e4ba4350 · outbound

This paper cites Neural machine translation of rare words with subword units.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Neural machine translation of rare words with subword units

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.638351Z

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.

Observation d77d4d9f-b228-43d4-a81b-eebd0581f86f · outbound

This paper cites Neural machine translation of rare words with subword units.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Neural machine translation of rare words with subword units

Reference 46

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.017589Z

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.

Observation 32fb769d-68dd-4722-a9bb-46039949b027 · outbound

This paper cites Skinnider.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Skinnider

Reference 47

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.790296Z

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.

Observation 2c067427-c970-4e4f-b6a9-df28c2c1f67f · outbound

This paper cites COCONUT online: Collection of open natural products database.Journal of Cheminformatics, 13(1):2, 2021.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES COCONUT online: Collection of open natural products database.Journal of Cheminformatics, 13(1):2, 2021

Reference 48

Resolution: malformed identifier; raw_fallback, observed 2026-07-11T03:47:52.664769Z

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.

Observation 2d911181-81a0-4a6c-8a63-96f29dfeb033 · outbound

This paper cites Linguistic laws meet protein sequences: A comparative analysis of subword tokenization methods, 2024.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Linguistic laws meet protein sequences: A comparative analysis of subword tokenization methods, 2024

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-07-11T03:47:52.689876Z

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.

Observation 93329e57-421c-4892-b768-bea2048311f7 · outbound

This paper cites Ülgen, Nilgün Karalı, and Arzucan Özgür.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Ülgen, Nilgün Karalı, and Arzucan Özgür

Reference 50

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.901649Z

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.

Observation ab4828f2-b58f-4b8e-86b6-57ef7ca03d99 · outbound

This paper cites Tingle, Khanh G.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Tingle, Khanh G

Reference 51

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.957064Z

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.

Observation bafcdac0-f60d-43b5-b7f4-c299d951637f · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES LLaMA: Open and Efficient Foundation Language Models

Reference 52

Resolution: verified exact; local_arxiv, observed 2026-07-11T03:47:48.527944Z

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.

Observation 33871cd9-2fbf-42da-9e9f-2a38ce5aa8f9 · outbound

This paper cites Ucak, Islambek Ashyrmamatov, and Juyong Lee.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Ucak, Islambek Ashyrmamatov, and Juyong Lee

Reference 53

Resolution: malformed identifier; raw_fallback, observed 2026-07-11T03:47:52.591354Z

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.

Observation ac076060-408f-4c26-9f84-cf9a135f6171 · outbound

This paper cites Tokenization for molecular foundation models.Journal of Chemical Information and Modeling, 66(3):1384–1393, 2026.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Tokenization for molecular foundation models.Journal of Chemical Information and Modeling, 66(3):1384–1393, 2026

Reference 54

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.095034Z

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.

Observation d44610aa-489a-4383-925b-1f2e96ed8e7d · outbound

This paper cites In: Che, W., Nabende, J., Shutova, E., Pilehvar, M.T.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES In: Che, W., Nabende, J., Shutova, E., Pilehvar, M.T

Reference 55

Resolution: metadata mismatch; doi, observed 2026-07-11T03:47:46.722668Z

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.

Observation 976f5da0-ca25-4135-8a82-e9e48fdca348 · outbound

This paper cites SMILES, a chemical language and information system.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES SMILES, a chemical language and information system

Reference 56

Resolution: malformed identifier; raw_fallback, observed 2026-07-11T03:47:52.781862Z

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.

Observation 5dd37a15-1741-4077-b6c7-ec15b1cc6295 · outbound

This paper cites Feinberg, Joseph Gomes, Caleb Geniesse, Aneesh S.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Feinberg, Joseph Gomes, Caleb Geniesse, Aneesh S

Reference 57

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.547235Z

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.

Observation 0b4d4730-7f96-4d5a-abd5-2c41c69465c4 · outbound

This paper cites Manners, James Blackshaw, Sybilla Corbett, Marleen de Veij, Haris Ioannidis, David Mendez Lopez, Juan F.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Manners, James Blackshaw, Sybilla Corbett, Marleen de Veij, Haris Ioannidis, David Mendez Lopez, Juan F

Reference 58

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.515804Z

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.

Observation 49a3c7cb-578a-488f-b57f-ef9b53a37dcf · outbound

This paper cites Tokenization and the Noiseless Channel.

Where to cut, how deep: BPE and Unigram-LM on chemistry SMILES Tokenization and the Noiseless Channel

Reference 59

Resolution: verified exact; doi, observed 2026-07-11T03:47:46.460245Z

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.

Coverage vector

Reference resolution

Outbound references

Pith citing papers