Paper Citation Record · LEDGER

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR

As of 9 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 3 inbound Pith citation observations for arXiv:2505.22063.

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

pith.paper-citation-record.v1
2505.22063 v1

Coverage vector

measured 40 of 40 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T13:21:04.016514Z

measured 43 of 43 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+00:00

measured 3 of 3 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T13:20:59.116638Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-18T16:31:37.216325Z

Reference resolution

40 of 40 outbound references displayed

verified exact0
verified fuzzy24
unresolved15
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 2e613eff-e2e5-4f72-b4d8-6babb144bd55 · outbound

This paper cites Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-07T13:20:59.116638Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f67a0cf2-0941-4e63-8894-dcf0c455fb23 · outbound

This paper cites To achieve this goal, EThai-ASR first introduces a self-evolving data refine- ment strategy to improve the quality of weakly labeled datasets.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR To achieve this goal, EThai-ASR first introduces a self-evolving data refine- ment strategy to improve the quality of weakly labeled datasets

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:09.238768Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 8c4bea92-d872-4137-81eb-d7c470e68e7d · outbound

This paper cites Giga2 Test.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Giga2 Test

Reference 3

Resolution: malformed identifier; raw_fallback, observed 2026-08-07T13:21:09.103821Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5d54473f-cb49-4ff6-a855-959fc9abb6ec · outbound

This paper cites an unresolved cited work.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Unresolved cited work

Reference 4

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:21:08.947321Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 40e9f4bf-0adf-4b79-a859-7c02012fe5f6 · outbound

This paper cites GPT-4 Technical Report.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR GPT-4 Technical Report

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T13:20:59.655332Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8ebfb63-59ed-4b08-9817-7c8cc5a437a8 · outbound

This paper cites Enhanc- ing Speech Emotion Recognition Using Dual Feature Extraction Encoders,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Enhanc- ing Speech Emotion Recognition Using Dual Feature Extraction Encoders,

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:08.757011Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation c2175b7c-ec00-48a4-8af6-d6341da84a2e · outbound

This paper cites Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T13:20:59.906754Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5432cebb-568f-4a4d-81dc-ccbf021f0097 · outbound

This paper cites Qwen2-Audio Technical Report.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Qwen2-Audio Technical Report

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:00.052267Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f04f7ef6-2bcc-4e72-98e8-24e83a39dfdd · outbound

This paper cites SALMONN: Towards Generic Hearing Abilities for Large Language Models,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR SALMONN: Towards Generic Hearing Abilities for Large Language Models,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:08.560784Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 639351cb-5db9-4937-979d-1a2c9480b532 · outbound

This paper cites Meta- Adaptable-Adapter: Efficient adaptation of self-supervised mod- els for low-resource speech recognition,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Meta- Adaptable-Adapter: Efficient adaptation of self-supervised mod- els for low-resource speech recognition,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:08.349908Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 2793e49e-c046-472f-a088-6aa1901b3705 · outbound

This paper cites Stable Distilla- tion: Regularizing Continued Pre-Training for Low-Resource Au- tomatic Speech Recognition,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Stable Distilla- tion: Regularizing Continued Pre-Training for Low-Resource Au- tomatic Speech Recognition,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:08.162749Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 6d7c5e71-4c56-43b3-8173-a52ba883ae51 · outbound

This paper cites Com- mon V oice: A Massively-Multilingual Speech Corpus,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Com- mon V oice: A Massively-Multilingual Speech Corpus,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.988828Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation dfa8aa0a-fe15-4bf8-937a-2f10d6cffc52 · outbound

This paper cites Lib- rispeech: An ASR corpus based on public domain audio books,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Lib- rispeech: An ASR corpus based on public domain audio books,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.846169Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 69136a8a-2d57-42f2-b874-42882b5ce5cc · outbound

This paper cites GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinement.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinement

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:00.909287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f1f23f26-7d2f-4b06-9743-b0ed97943858 · outbound

This paper cites MSR-86K: An Evolving, Multilingual Corpus with 86,300 Hours of Transcribed Audio for Speech Recognition Research.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR MSR-86K: An Evolving, Multilingual Corpus with 86,300 Hours of Transcribed Audio for Speech Recognition Research

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:01.072501Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b2cf2b8-9d76-4137-9502-a74b60a3212a · outbound

This paper cites Thai-Dialect: Low Resource Thai Dialectal Speech to Text Cor- pora,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Thai-Dialect: Low Resource Thai Dialectal Speech to Text Cor- pora,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.665012Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation d1a1ca25-81d4-438e-98f1-289fa0e314a9 · outbound

This paper cites Automatic Speech Recognition Techniques for Transcription of Thai Traditional Medicine Texts,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Automatic Speech Recognition Techniques for Transcription of Thai Traditional Medicine Texts,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.557410Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 5da37c4f-3f5a-4040-9243-0b7a913ccf02 · outbound

This paper cites Comprehensive Benchmarking and Analysis of Open Pretrained Thai Speech Recognition Models,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Comprehensive Benchmarking and Analysis of Open Pretrained Thai Speech Recognition Models,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.386183Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation a30b3edd-b92d-48ef-aaad-76f0cd9d70a5 · outbound

This paper cites Typhoon: Thai Large Language Models.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Typhoon: Thai Large Language Models

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:01.632847Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0305903d-1d7a-47b6-822d-14b70bec697e · outbound

This paper cites Typhoon 2: A Family of Open Text and Multimodal Thai Large Language Models.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Typhoon 2: A Family of Open Text and Multimodal Thai Large Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:01.788931Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b4e8642d-3e79-4e28-b2cc-48be220962f8 · outbound

This paper cites Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Seed-ASR: Understanding Diverse Speech and Contexts with LLM-based Speech Recognition

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:01.949770Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 80938492-0b5f-4509-a07f-c0d9847afa3b · outbound

This paper cites AudioPaLM: A Large Language Model That Can Speak and Listen.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR AudioPaLM: A Large Language Model That Can Speak and Listen

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:02.086190Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f84b21eb-086a-4199-95f3-c3341fda2126 · outbound

This paper cites Enhancing Low-Resource Language and Instruction Following Capabilities of Audio Language Models.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Enhancing Low-Resource Language and Instruction Following Capabilities of Audio Language Models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:02.170670Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 218d59e2-48d6-42c7-868d-a8db7e2cc250 · outbound

This paper cites Skipformer: A Skip-and-Recover Strategy for Efficient Speech Recognition,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Skipformer: A Skip-and-Recover Strategy for Efficient Speech Recognition,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.249280Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation afcc574f-b1f7-4dba-b237-6c2a9c82bb34 · outbound

This paper cites Zipformer: A faster and better encoder for automatic speech recognition,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Zipformer: A faster and better encoder for automatic speech recognition,

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:07.045956Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f30be208-10f8-43f9-9ee5-7adc9555de31 · outbound

This paper cites CosyAudio: Improving Audio Generation with Confidence Scores and Synthetic Captions.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR CosyAudio: Improving Audio Generation with Confidence Scores and Synthetic Captions

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:02.529518Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 83b43c1f-d358-4fc6-9a88-ac98d5f6b7bf · outbound

This paper cites scPLAN: a hierar- chical computational framework for single transcriptomics data annotation, integration and cell-type label refinement,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR scPLAN: a hierar- chical computational framework for single transcriptomics data annotation, integration and cell-type label refinement,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:06.886505Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation d8a367e0-d5b1-4aae-bfb5-8ddbe93e6733 · outbound

This paper cites Data driven multi-objective economic-environmental robust optimiza- tion for refinery planning with multiple modes under uncertainty,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Data driven multi-objective economic-environmental robust optimiza- tion for refinery planning with multiple modes under uncertainty,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:06.698799Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation a8034192-82e4-46e5-8158-459ea2fa92f9 · outbound

This paper cites A survey of dataset refinement for problems in computer vision datasets,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR A survey of dataset refinement for problems in computer vision datasets,

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:06.463276Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 7e5f6173-1082-4bb1-a63e-43ac62080280 · outbound

This paper cites A multimodal cross- transformer-based model to predict mild cognitive impairment us- ing speech, language and vision,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR A multimodal cross- transformer-based model to predict mild cognitive impairment us- ing speech, language and vision,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:06.183414Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 8fe28523-c19c-4845-bb19-6ac679bf5d49 · outbound

This paper cites Qformer: An efficient quaternion transformer for image denoising,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Qformer: An efficient quaternion transformer for image denoising,

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:05.941077Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation f703c349-e2c1-4198-bd3d-29e09f5df4f1 · outbound

This paper cites Unveiling the Potential of LLM-Based ASR on Chinese Open-Source Datasets,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Unveiling the Potential of LLM-Based ASR on Chinese Open-Source Datasets,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:05.749311Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 4102aef1-9138-4fa2-9d93-5a155ee36f03 · outbound

This paper cites LoRA: Low-Rank Adaptation of Large Language Models,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR LoRA: Low-Rank Adaptation of Large Language Models,

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:05.487002Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 74f4251d-86d6-4d7d-be6f-47d3e70c558f · outbound

This paper cites Lora+: Efficient low rank adap- tation of large models,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Lora+: Efficient low rank adap- tation of large models,

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:05.256147Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 2df6c57d-44dc-41d4-bc42-e613dccb7871 · outbound

This paper cites Sequence-level Knowledge Distillation for Model Compression of Attention- based Sequence-to-sequence Speech Recognition,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Sequence-level Knowledge Distillation for Model Compression of Attention- based Sequence-to-sequence Speech Recognition,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:05.087310Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation de8a8f79-bd61-4c91-95bf-43f606a5fb33 · outbound

This paper cites Once-for-all sequence compres- sion for self-supervised speech models,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Once-for-all sequence compres- sion for self-supervised speech models,

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:04.927255Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation 10fcdc6b-8746-4d61-9a09-7869d035ac67 · outbound

This paper cites Visionzip: Longer is better but not necessary in vision language models,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Visionzip: Longer is better but not necessary in vision language models,

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:03.600207Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4b8a6322-451b-4ed4-ad9c-d2caa6fce31d · outbound

This paper cites VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-07T13:21:03.727067Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 15b77c1b-bfd2-4212-81ce-5ff6bf1f68f8 · outbound

This paper cites FLEURS: FEW-Shot Learning Evaluation of Universal Representations of Speech,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR FLEURS: FEW-Shot Learning Evaluation of Universal Representations of Speech,

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:04.788942Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Observation a738b125-40d7-43a3-92c5-2f3607b0ce9f · outbound

This paper cites Robust Speech Recognition via Large-Scale Weak Supervision,.

Weakly Supervised Data Refinement and Flexible Sequence Compression for Efficient Thai LLM-based ASR Robust Speech Recognition via Large-Scale Weak Supervision,

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:21:04.610950Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

Pith citing papers

Observation 2e613eff-e2e5-4f72-b4d8-6babb144bd55 · inbound