Paper Citation Record · LEDGER

Step-Audio 2 Technical Report

As of 14 July 2026, this Paper Citation Record lists 84 of 84 outbound references and 50 inbound Pith citation observations for arXiv:2507.16632.

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

pith.paper-citation-record.v1
2507.16632 v3

Coverage vector

measured 84 of 84 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-16T05:59:50.900436Z

measured 134 of 134 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache

measured 50 of 50 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-03T15:17:52.966144Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-07-04T13:59:52.899067Z

Reference resolution

84 of 84 outbound references displayed

verified exact51
verified fuzzy17
unresolved1
parse uncertain0
malformed identifier2
metadata mismatch13

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation f313ab2b-dc9a-4109-a0a2-f5ea84ba8cf5 · outbound

This paper cites Seed-TTS: A Family of High-Quality Versatile Speech Generation Models.

Step-Audio 2 Technical Report Seed-TTS: A Family of High-Quality Versatile Speech Generation Models

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.008000Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 584c11ac-e0e1-4b6e-9d8b-5074bea8d36a · outbound

This paper cites PaLM 2 Technical Report.

Step-Audio 2 Technical Report PaLM 2 Technical Report

Reference 2

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.041600Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 14197ef2-d9f1-47ad-9539-a774f9e0ac71 · outbound

This paper cites wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations.

Step-Audio 2 Technical Report wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations

Reference 3

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.046218Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 5a2ba534-d99c-4f44-a1a8-279508c070d9 · outbound

This paper cites Qwen Technical Report.

Step-Audio 2 Technical Report Qwen Technical Report

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.049209Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation eb5c602e-02af-4a2b-a973-a8bddc81995d · outbound

This paper cites Seed-asr: Understanding diverse speech and contexts with llm-based speech recognition.

Step-Audio 2 Technical Report Seed-asr: Understanding diverse speech and contexts with llm-based speech recognition

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.053374Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation b830277d-a5b9-48e4-8861-901a9974e525 · outbound

This paper cites Better speech synthesis through scaling.

Step-Audio 2 Technical Report Better speech synthesis through scaling

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:50.966924Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 02717722-fb45-4ff4-9c97-056e1c3b70d5 · outbound

This paper cites Audiolm: a language modeling approach to audio generation.

Step-Audio 2 Technical Report Audiolm: a language modeling approach to audio generation

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.254155Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation e84d293d-331f-4fb7-92ca-488ef7ab9a37 · outbound

This paper cites GigaSpeech: An Evolving, Multi-Domain ASR Corpus with 10,000 Hours of Transcribed Audio.

Step-Audio 2 Technical Report GigaSpeech: An Evolving, Multi-Domain ASR Corpus with 10,000 Hours of Transcribed Audio

Reference 8

Resolution: malformed identifier; doi_truncated, observed 2026-05-16T05:59:50.962700Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 6665ef24-3b40-480a-bf0d-30f4124a50d6 · outbound

This paper cites Chen, Y.

Step-Audio 2 Technical Report Chen, Y

Reference 9

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:50.972489Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation cc072def-b9ec-4f42-bd4c-f70984df4af5 · outbound

This paper cites BEATs: Audio Pre-Training with Acoustic Tokenizers.

Step-Audio 2 Technical Report BEATs: Audio Pre-Training with Acoustic Tokenizers

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:50.976241Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 21ba5f54-8191-4fb7-9a91-5db9bf3faa18 · outbound

This paper cites neurips.cc/paper/2020/hash/1457c0d6bfcb4967418bfb8ac142f64a-Abstract.html.

Step-Audio 2 Technical Report neurips.cc/paper/2020/hash/1457c0d6bfcb4967418bfb8ac142f64a-Abstract.html

Reference 11

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:50.942128Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a85c5273-a4ad-4bb3-83da-cfd9fe8cb236 · outbound

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

Step-Audio 2 Technical Report Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:50.980061Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation eaa52168-ab95-440f-a5e4-5fb072d9d9b2 · outbound

This paper cites Qwen2-Audio Technical Report.

Step-Audio 2 Technical Report Qwen2-Audio Technical Report

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:50.983298Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation e8bdfe38-b2f6-4217-95aa-0588bd977870 · outbound

This paper cites Strongly Recommend Advancing.

Step-Audio 2 Technical Report Strongly Recommend Advancing

Reference 14

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:50.987583Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 63b15d25-308f-4b49-a0fb-40412b34409c · outbound

This paper cites High Fidelity Neural Audio Compression.

Step-Audio 2 Technical Report High Fidelity Neural Audio Compression

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:50.991544Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 5666ce06-555b-459c-a795-3e1c3066e572 · outbound

This paper cites Moshi: a speech-text foundation model for real-time dialogue.

Step-Audio 2 Technical Report Moshi: a speech-text foundation model for real-time dialogue

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:50.994640Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 0d626368-8034-42aa-a316-fabdd8107893 · outbound

This paper cites Pengi: An audio language model for audio tasks.

Step-Audio 2 Technical Report Pengi: An audio language model for audio tasks

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.225299Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 6f8fe792-902b-40f3-abc1-6e51ac228096 · outbound

This paper cites Kimi-Audio Technical Report.

Step-Audio 2 Technical Report Kimi-Audio Technical Report

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:50.998210Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 9baed511-0246-4cd0-ade7-9a5f45a3258f · outbound

This paper cites CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models.

Step-Audio 2 Technical Report CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models

Reference 19

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.001250Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 55cc6923-acb3-4e47-9d9f-32efb0032134 · outbound

This paper cites CosyVoice: A Scalable Multilingual Zero-shot Text-to-speech Synthesizer based on Supervised Semantic Tokens.

Step-Audio 2 Technical Report CosyVoice: A Scalable Multilingual Zero-shot Text-to-speech Synthesizer based on Supervised Semantic Tokens

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.004746Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation cdb8ac00-ddad-47a7-abd9-6780ddf68f6e · outbound

This paper cites Llama-omni: Seamless speech interaction with large language models.

Step-Audio 2 Technical Report Llama-omni: Seamless speech interaction with large language models

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.023035Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 31cc283f-31f1-437c-b415-677956b5a359 · outbound

This paper cites Shao, H.

Step-Audio 2 Technical Report Shao, H

Reference 22

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.011598Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 3df05a25-16d9-46bc-b7f0-e790b6d4b4bc · outbound

This paper cites Gemmeke, Daniel P.

Step-Audio 2 Technical Report Gemmeke, Daniel P

Reference 23

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:50.946828Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation cc18c3e4-6860-40e5-82f7-e582062a4df1 · outbound

This paper cites Audio flamingo 2: An audio- language model with long-audio understanding and expert rea- soning abilities.

Step-Audio 2 Technical Report Audio flamingo 2: An audio- language model with long-audio understanding and expert rea- soning abilities

Reference 24

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.015617Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 82669d45-d189-4502-9bf1-235dee64ab06 · outbound

This paper cites Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models.

Step-Audio 2 Technical Report Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models

Reference 25

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.018782Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation d7c2584d-c11b-4f00-8334-0e1eaf21578e · outbound

This paper cites author Zhou, A.

Step-Audio 2 Technical Report author Zhou, A

Reference 26

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:50.937574Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation d69f09c5-d35b-4212-ae7d-d14123ac7dd9 · outbound

This paper cites Joint audio and speech understanding.

Step-Audio 2 Technical Report Joint audio and speech understanding

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.228824Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 74ecce0c-3ad2-4bb4-8d8d-44359d549937 · outbound

This paper cites Liu, Leonid Karlinsky, and James Glass.

Step-Audio 2 Technical Report Liu, Leonid Karlinsky, and James Glass

Reference 28

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.125480Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation f756621c-c9f1-4f1c-afcf-4a1e420c5848 · outbound

This paper cites The Llama 3 Herd of Models.

Step-Audio 2 Technical Report The Llama 3 Herd of Models

Reference 29

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.128823Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 986ec86d-5bf2-489e-a7fd-de4c8516c67d · outbound

This paper cites HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units.

Step-Audio 2 Technical Report HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units

Reference 30

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.132301Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 77ab2a72-895c-410e-9eb3-653860b0be7d · outbound

This paper cites Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model.

Step-Audio 2 Technical Report Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.136199Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a90d7b25-65a4-4383-92f1-9f8da6f8cceb · outbound

This paper cites Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction.

Step-Audio 2 Technical Report Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-05-18T13:39:48.481060Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation f5b1eb26-4ab8-4809-a2a7-996229daaf44 · outbound

This paper cites Audiogpt: Understanding and generating speech, music, sound, and talking head.

Step-Audio 2 Technical Report Audiogpt: Understanding and generating speech, music, sound, and talking head

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.248005Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation d105612e-3007-4426-9484-d29d2fef501c · outbound

This paper cites GPT-4o System Card.

Step-Audio 2 Technical Report GPT-4o System Card

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.144069Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation bc532546-80a3-41ae-9c31-dc66440a2370 · outbound

This paper cites CochlScene: Acquisition of acoustic scene data using crowdsourcing.

Step-Audio 2 Technical Report CochlScene: Acquisition of acoustic scene data using crowdsourcing

Reference 35

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:50.958649Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 44fe38c8-885e-4938-85e1-12168b10b1b6 · outbound

This paper cites Wavtokenizer: an efficient acoustic discrete codec tokenizer for audio language modeling.

Step-Audio 2 Technical Report Wavtokenizer: an efficient acoustic discrete codec tokenizer for audio language modeling

Reference 36

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.147949Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation e3007030-51fa-4b62-9b2f-55f11a1d4d9d · outbound

This paper cites CVSS Corpus and Massively Multilingual Speech-to-Speech Translation.

Step-Audio 2 Technical Report CVSS Corpus and Massively Multilingual Speech-to-Speech Translation

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.151543Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation f731bf00-c1e8-4fdf-b0f2-6c0625c2fdc1 · outbound

This paper cites Direct speech-to-speech translation with a sequence-to-sequence model.

Step-Audio 2 Technical Report Direct speech-to-speech translation with a sequence-to-sequence model

Reference 38

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.154816Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 608c5477-4cee-4cf6-935c-3213309a5b34 · outbound

This paper cites Translatotron 2: High-quality direct speech-to-speech translation with voice preservation.

Step-Audio 2 Technical Report Translatotron 2: High-quality direct speech-to-speech translation with voice preservation

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.266889Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation f3d23c3f-2582-4a81-94b3-58743b50a273 · outbound

This paper cites Speak, Read and Prompt: High-Fidelity Text-to-Speech with Minimal Supervision.

Step-Audio 2 Technical Report Speak, Read and Prompt: High-Fidelity Text-to-Speech with Minimal Supervision

Reference 40

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.160129Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 2d7bedba-3763-4b18-b252-7d0482bebcc3 · outbound

This paper cites Audiocaps: Generating captions for audios in the wild.

Step-Audio 2 Technical Report Audiocaps: Generating captions for audios in the wild

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.272307Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation dca4162f-ba8b-4bd4-9128-75eaac40980d · outbound

This paper cites HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis.

Step-Audio 2 Technical Report HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis

Reference 42

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.164871Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation d93b5c71-6bd4-4347-855d-f0dbd4dda283 · outbound

This paper cites Audio flamingo: A novel audio language model with few-shot learning and dialogue abilities.

Step-Audio 2 Technical Report Audio flamingo: A novel audio language model with few-shot learning and dialogue abilities

Reference 43

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.170112Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4dba186c-102d-4fe6-afa9-fee767f8797e · outbound

This paper cites Transvip: Speech to speech translation system with voice and isochrony preservation.

Step-Audio 2 Technical Report Transvip: Speech to speech translation system with voice and isochrony preservation

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.231754Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation fcdc650e-57f5-452a-9f2a-c29f431e5e03 · outbound

This paper cites Textless Speech-to-Speech Translation on Real Data.

Step-Audio 2 Technical Report Textless Speech-to-Speech Translation on Real Data

Reference 45

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.175460Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 604528b6-756c-4732-9d83-2299a1a9d949 · outbound

This paper cites BigVGAN: A Universal Neural Vocoder with Large-Scale Training.

Step-Audio 2 Technical Report BigVGAN: A Universal Neural Vocoder with Large-Scale Training

Reference 46

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.180223Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 49e7a81f-9f7f-425b-9f8d-373ffda09316 · outbound

This paper cites Advancing large language models to capture varied speaking styles and respond properly in spoken conversations.

Step-Audio 2 Technical Report Advancing large language models to capture varied speaking styles and respond properly in spoken conversations

Reference 47

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.184881Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 3256bd3c-b217-40e0-8fe0-c67b35a120bb · outbound

This paper cites Paralinguistics-enhanced large language modeling of spoken dialogue.

Step-Audio 2 Technical Report Paralinguistics-enhanced large language modeling of spoken dialogue

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.244677Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4d636086-43ce-4dac-a310-bc02fbed353e · outbound

This paper cites Costa-jussà, Maha Elbayad, Sravya Popuri, Paul-Ambroise Duquenne, Robin Algayres, Ruslan Mavlyutov, Itai Gat, et al.

Step-Audio 2 Technical Report Costa-jussà, Maha Elbayad, Sravya Popuri, Paul-Ambroise Duquenne, Robin Algayres, Ruslan Mavlyutov, Itai Gat, et al

Reference 49

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.189813Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4aa8388f-57eb-47ca-ae64-9f7d90a7e7b1 · outbound

This paper cites GPT-4 Technical Report.

Step-Audio 2 Technical Report GPT-4 Technical Report

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.257245Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 7659cef7-b205-41c5-9142-77a884ff3ba2 · outbound

This paper cites Introducing ChatGPT.

Step-Audio 2 Technical Report Introducing ChatGPT

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.260503Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 456fb515-31ed-4151-b8b9-5df49f867b1d · outbound

This paper cites Deep voice 3: 2000-speaker neural text-to-speech.

Step-Audio 2 Technical Report Deep voice 3: 2000-speaker neural text-to-speech

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.263364Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a9a3228c-6a42-4fa2-89ea-ef42d1c19485 · outbound

This paper cites Robust speech recognition via large-scale weak supervision.

Step-Audio 2 Technical Report Robust speech recognition via large-scale weak supervision

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.269653Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation fe1f1933-5336-456b-b13f-c4804874ae76 · outbound

This paper cites Direct preference optimization: Your language model is secretly a reward model.

Step-Audio 2 Technical Report Direct preference optimization: Your language model is secretly a reward model

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.275004Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 133feca1-a02c-4ef2-a45a-904b8aa3e946 · outbound

This paper cites FastSpeech 2: Fast and High-Quality End-to-End Text to Speech.

Step-Audio 2 Technical Report FastSpeech 2: Fast and High-Quality End-to-End Text to Speech

Reference 55

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.194912Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation be9a0682-e75b-4e39-af23-69b472953c73 · outbound

This paper cites Omni-r1: Do you really need audio to fine-tune your audio llm?.

Step-Audio 2 Technical Report Omni-r1: Do you really need audio to fine-tune your audio llm?

Reference 56

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.199582Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation e14ce1df-f34b-45f9-8f90-b6a18f7055aa · outbound

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

Step-Audio 2 Technical Report AudioPaLM: A Large Language Model That Can Speak and Listen

Reference 57

Resolution: verified exact; arxiv_id, observed 2026-05-16T07:07:57.965390Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 402c49bb-5830-4b98-8bd2-bbc9940d3ea2 · outbound

This paper cites MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark.

Step-Audio 2 Technical Report MMAU: A Massive Multi-Task Audio Understanding and Reasoning Benchmark

Reference 58

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.207537Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a73a0bd6-2198-4ce1-b004-1003b080d9a6 · outbound

This paper cites Natural tts synthesis by conditioning wavenet on mel spectrogram predictions.

Step-Audio 2 Technical Report Natural tts synthesis by conditioning wavenet on mel spectrogram predictions

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.250651Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4239f7c3-949a-4f5c-9cc3-6c927f9bd808 · outbound

This paper cites Snac: Multi-scale neural audio codec.arXiv preprint arXiv:2410.14411.

Step-Audio 2 Technical Report Snac: Multi-scale neural audio codec.arXiv preprint arXiv:2410.14411

Reference 60

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.212358Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 8fa69d21-6b82-46e8-94d1-9063310736b8 · outbound

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

Step-Audio 2 Technical Report SALMONN: Towards Generic Hearing Abilities for Large Language Models

Reference 61

Resolution: verified exact; arxiv_id, observed 2026-05-18T02:29:46.683133Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation bfff5da5-5941-43f4-880e-be978cb54d63 · outbound

This paper cites Springer Science & Business Media.

Step-Audio 2 Technical Report Springer Science & Business Media

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.238224Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 9533a609-8b97-494b-8d2e-de0f94ebee15 · outbound

This paper cites an unresolved cited work.

Step-Audio 2 Technical Report Unresolved cited work

Reference 63

Resolution: unresolved; raw_fallback, observed 2026-05-16T05:59:51.241703Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 6c55eaed-62d2-4b4d-ab26-3cfa165e6684 · outbound

This paper cites CoVoST 2 and Massively Multilingual Speech-to-Text Translation.

Step-Audio 2 Technical Report CoVoST 2 and Massively Multilingual Speech-to-Text Translation

Reference 64

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.222191Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 48038e2d-9ee6-464e-985b-d4c9e5d0aff9 · outbound

This paper cites Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers.

Step-Audio 2 Technical Report Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers

Reference 65

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.056488Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a925d1f3-84b8-455a-9fad-945bdf949598 · outbound

This paper cites Spark-TTS: An Efficient LLM-Based Text-to-Speech Model with Single-Stream Decoupled Speech Tokens.

Step-Audio 2 Technical Report Spark-TTS: An Efficient LLM-Based Text-to-Speech Model with Single-Stream Decoupled Speech Tokens

Reference 66

Resolution: verified exact; arxiv_id, observed 2026-05-17T09:39:51.552824Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 532a21d1-7a9f-4c26-a582-7550813991d3 · outbound

This paper cites Freeze-omni: A smart and low latency speech-to-speech dialogue model with frozen llm.

Step-Audio 2 Technical Report Freeze-omni: A smart and low latency speech-to-speech dialogue model with frozen llm

Reference 67

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.063110Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 63183282-1e75-44c9-94dd-9a9f6e44dfa2 · outbound

This paper cites Maskgct: Zero-shot text- to-speech with masked generative codec transformer.

Step-Audio 2 Technical Report Maskgct: Zero-shot text- to-speech with masked generative codec transformer

Reference 68

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.066886Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 21167f65-a8a9-47a6-a255-0038178564b5 · outbound

This paper cites Tacotron: Towards End-to-End Speech Synthesis.

Step-Audio 2 Technical Report Tacotron: Towards End-to-End Speech Synthesis

Reference 69

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.070297Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 54720965-253d-434f-b47f-9363e0f5202d · outbound

This paper cites Finetuned Language Models Are Zero-Shot Learners.

Step-Audio 2 Technical Report Finetuned Language Models Are Zero-Shot Learners

Reference 70

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.073929Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 9066e9cc-b480-41d6-9c88-98dfbc484dbc · outbound

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

Step-Audio 2 Technical Report Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation

Reference 71

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.078240Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation ff6958e9-3557-478e-b356-2c5656001576 · outbound

This paper cites Mini-omni: Language models can hear, talk while thinking in streaming.

Step-Audio 2 Technical Report Mini-omni: Language models can hear, talk while thinking in streaming

Reference 72

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.083165Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation c6df6a3f-034e-4b1a-b016-83e6fa094b00 · outbound

This paper cites Mini-omni2: Towards open-source gpt-4o with vision, speech and duplex capabilities.

Step-Audio 2 Technical Report Mini-omni2: Towards open-source gpt-4o with vision, speech and duplex capabilities

Reference 73

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.088260Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 2a28d9ff-c3c3-4c9e-909b-477a118883d6 · outbound

This paper cites Bigcodec: Pushing the limits of low-bitrate neural speech codec.

Step-Audio 2 Technical Report Bigcodec: Pushing the limits of low-bitrate neural speech codec

Reference 74

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.092542Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 0cacf5bd-6dee-4377-aee6-6cfa7a27593b · outbound

This paper cites Qwen2.5-Omni Technical Report.

Step-Audio 2 Technical Report Qwen2.5-Omni Technical Report

Reference 75

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.096121Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 106fd36e-5574-4659-a943-269086e2d72c · outbound

This paper cites Chih-Kai Yang, Yu-Kuan Fu, Chen-An Li, Yi-Cheng Lin, Yu-Xiang Lin, Wei-Chih Chen, Ho Lam Chung, Chun-Yi Kuan, Wei-Ping Huang, Ke-Han Lu, and 1 others.

Step-Audio 2 Technical Report Chih-Kai Yang, Yu-Kuan Fu, Chen-An Li, Yi-Cheng Lin, Yu-Xiang Lin, Wei-Chih Chen, Ho Lam Chung, Chun-Yi Kuan, Wei-Ping Huang, Ke-Han Lu, and 1 others

Reference 76

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.100939Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 0debbe78-392b-4e71-9905-98b729fdb6ec · outbound

This paper cites Soundstream: An end-to-end neural audio codec.

Step-Audio 2 Technical Report Soundstream: An end-to-end neural audio codec

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.235067Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation c1b2a798-1d33-450d-853d-7e9925e973bc · outbound

This paper cites GLM-4-Voice: Towards Intelligent and Human-Like End-to-End Spoken Chatbot.

Step-Audio 2 Technical Report GLM-4-Voice: Towards Intelligent and Human-Like End-to-End Spoken Chatbot

Reference 78

Resolution: verified exact; local_arxiv, observed 2026-05-16T05:59:51.104219Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 95605eb4-5fa4-4eb7-be91-052eb72ed527 · outbound

This paper cites WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition.

Step-Audio 2 Technical Report WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition

Reference 79

Resolution: verified fuzzy; raw_fallback, observed 2026-05-16T05:59:51.277513Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 0fdc7dad-ae15-41b2-b68d-4684f58df13e · outbound

This paper cites WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition.

Step-Audio 2 Technical Report WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition

Reference 80

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.107662Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 5c555d22-4377-4e1e-b64f-0cc68308a95a · outbound

This paper cites Minimax- speech: Intrinsic zero-shot text-to-speech with a learnable speaker encoder.

Step-Audio 2 Technical Report Minimax- speech: Intrinsic zero-shot text-to-speech with a learnable speaker encoder

Reference 81

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.111664Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 9529745b-1880-4133-a879-50076ef4bab5 · outbound

This paper cites SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities.

Step-Audio 2 Technical Report SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities

Reference 82

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.115148Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation b7dc5c6a-1977-4d51-bf91-c5efbf7151ac · outbound

This paper cites Distinctive Feature Codec: Adaptive Segmentation for Efficient Speech Representation.

Step-Audio 2 Technical Report Distinctive Feature Codec: Adaptive Segmentation for Efficient Speech Representation

Reference 83

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.118689Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 2e2cc775-94ec-4211-bae6-f0d509600162 · outbound

This paper cites Speechtokenizer: Unified speech tokenizer for speech large language models.

Step-Audio 2 Technical Report Speechtokenizer: Unified speech tokenizer for speech large language models

Reference 84

Resolution: malformed identifier; arxiv_id, observed 2026-05-16T05:59:51.122199Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Pith citing papers

Observation 6967a63e-7cb4-44d3-8ffe-27fd0ef710a7 · inbound

StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs cites this paper.

StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs Step-Audio 2 Technical Report

Reference 76

Resolution: verified exact; local_arxiv, observed 2026-05-18T13:01:24.338010Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 81077e48-7d92-40ce-b112-33ab17992aa8 · inbound

Mind-Paced Speaking: A Dual-Brain Approach to Real-Time Reasoning in Spoken Language Models cites this paper.

Mind-Paced Speaking: A Dual-Brain Approach to Real-Time Reasoning in Spoken Language Models Step-Audio 2 Technical Report

Reference 2

Resolution: verified exact; local_arxiv, observed 2026-05-18T07:46:03.604065Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 147da74f-f87c-4c8b-9039-e73dec0e64b2 · inbound

Style Amnesia: Investigating Speaking Style Degradation and Mitigation in Multi-Turn Spoken Language Models cites this paper.

Style Amnesia: Investigating Speaking Style Degradation and Mitigation in Multi-Turn Spoken Language Models Step-Audio 2 Technical Report

Reference 44

Resolution: verified exact; local_arxiv, observed 2026-05-16T19:28:20.146197Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a17c5fd6-b258-46c9-92d0-c5c8a00e1fa8 · inbound

MCGA: A Multi-task Classical Chinese Literary Genre Audio Corpus cites this paper.

MCGA: A Multi-task Classical Chinese Literary Genre Audio Corpus Step-Audio 2 Technical Report

Reference 14

Resolution: metadata mismatch; local_arxiv, observed 2026-05-16T14:31:01.935898Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 996aed95-25fb-47e3-9b14-a39f84d42d38 · inbound

OmniFysics: Towards Physical Intelligence Evolution via Omni-Modal Signal Processing and Network Optimization cites this paper.

OmniFysics: Towards Physical Intelligence Evolution via Omni-Modal Signal Processing and Network Optimization Step-Audio 2 Technical Report

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-05-16T07:10:43.180020Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation f04c02fe-7319-430a-b7f0-1d45260bcec3 · inbound

TiCo: Time-Controllable Spoken Dialogue Model cites this paper.

TiCo: Time-Controllable Spoken Dialogue Model Step-Audio 2 Technical Report

Reference 69

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation af84ae87-827b-43bf-b89e-d701ca56c950 · inbound

Speaker-Reasoner: Scaling Interaction Turns and Reasoning Patterns for Timestamped Speaker-Attributed ASR cites this paper.

Speaker-Reasoner: Scaling Interaction Turns and Reasoning Patterns for Timestamped Speaker-Attributed ASR Step-Audio 2 Technical Report

Reference 18

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation cbe6cfe8-f1c6-473c-a28c-a274b5fbc96f · inbound

Rethinking Entropy Allocation in LLM-based ASR: Understanding the Dynamics between Speech Encoders and LLMs cites this paper.

Rethinking Entropy Allocation in LLM-based ASR: Understanding the Dynamics between Speech Encoders and LLMs Step-Audio 2 Technical Report

Reference 24

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 8b0dc400-a0cc-45ba-90dd-3bf7bccd5d8b · inbound

CapTalk: Unified Voice Design for Single-Utterance and Dialogue Speech Generation cites this paper.

CapTalk: Unified Voice Design for Single-Utterance and Dialogue Speech Generation Step-Audio 2 Technical Report

Reference 41

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation c8362045-2814-42ae-8e9d-f513bd817c25 · inbound

HumDial-EIBench: A Human-Recorded Multi-Turn Emotional Intelligence Benchmark for Audio Language Models cites this paper.

HumDial-EIBench: A Human-Recorded Multi-Turn Emotional Intelligence Benchmark for Audio Language Models Step-Audio 2 Technical Report

Reference 36

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4fd28479-650d-450f-a12f-1706decb1155 · inbound

Why Your Tokenizer Fails in Information Fusion: A Timing-Aware Pre-Quantization Fusion for Video-Enhanced Audio Tokenization cites this paper.

Why Your Tokenizer Fails in Information Fusion: A Timing-Aware Pre-Quantization Fusion for Video-Enhanced Audio Tokenization Step-Audio 2 Technical Report

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation c0227306-c1aa-4748-8363-252d69efc8f7 · inbound

Audio-Cogito: Towards Deep Audio Reasoning in Large Audio Language Models cites this paper.

Audio-Cogito: Towards Deep Audio Reasoning in Large Audio Language Models Step-Audio 2 Technical Report

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation e3bdb5eb-024e-4fb1-9ad3-bab3fface9a2 · inbound

VIBE: Voice-Induced open-ended Bias Evaluation for Large Audio-Language Models via Real-World Speech cites this paper.

VIBE: Voice-Induced open-ended Bias Evaluation for Large Audio-Language Models via Real-World Speech Step-Audio 2 Technical Report

Reference 47

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 098a68c3-d4a8-4817-93ca-beb400d52f7a · inbound

NIM4-ASR: Towards Efficient, Robust, and Customizable Real-Time LLM-Based ASR cites this paper.

NIM4-ASR: Towards Efficient, Robust, and Customizable Real-Time LLM-Based ASR Step-Audio 2 Technical Report

Reference 25

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a18e10ed-86e9-4ad4-8171-5bc692c015a4 · inbound

Audio-DeepThinker: Progressive Reasoning-Aware Reinforcement Learning for High-Quality Chain-of-Thought Emergence in Audio Language Models cites this paper.

Audio-DeepThinker: Progressive Reasoning-Aware Reinforcement Learning for High-Quality Chain-of-Thought Emergence in Audio Language Models Step-Audio 2 Technical Report

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation cfafa6b9-7fa4-4f79-9c18-f397e9f71339 · inbound

HalluAudio: A Comprehensive Benchmark for Hallucination Detection in Large Audio-Language Models cites this paper.

HalluAudio: A Comprehensive Benchmark for Hallucination Detection in Large Audio-Language Models Step-Audio 2 Technical Report

Reference 19

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 5ff5be2f-09d8-40a1-b9ce-a6766b63cf0c · inbound

SpeechParaling-Bench: A Comprehensive Benchmark for Paralinguistic-Aware Speech Generation cites this paper.

SpeechParaling-Bench: A Comprehensive Benchmark for Paralinguistic-Aware Speech Generation Step-Audio 2 Technical Report

Reference 30

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation fdf12d8b-4fd4-4979-99d3-43a59393846b · inbound

Audio2Tool: Speak, Call, Act -- A Dataset for Benchmarking Speech Tool Use cites this paper.

Audio2Tool: Speak, Call, Act -- A Dataset for Benchmarking Speech Tool Use Step-Audio 2 Technical Report

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation cf731bde-8186-4852-9a00-dc4c8ef4ba1c · inbound

Step-Audio-R1.5 Technical Report cites this paper.

Step-Audio-R1.5 Technical Report Step-Audio 2 Technical Report

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 74130fb8-fe7e-49f2-b93a-b983bf377927 · inbound

VocalParse: Towards Unified and Scalable Singing Voice Transcription with Large Audio Language Models cites this paper.

VocalParse: Towards Unified and Scalable Singing Voice Transcription with Large Audio Language Models Step-Audio 2 Technical Report

Reference 37

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 09770f2a-72e3-4ba9-9d48-d809e73a3b67 · inbound

How Should LLMs Listen While Speaking? A Study of User-Stream Routing in Full-Duplex Spoken Dialogue cites this paper.

How Should LLMs Listen While Speaking? A Study of User-Stream Routing in Full-Duplex Spoken Dialogue Step-Audio 2 Technical Report

Reference 43

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 88fed131-320d-463d-9be7-42bdf97521c2 · inbound

Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation cites this paper.

Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation Step-Audio 2 Technical Report

Reference 24

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation c37b2dcd-6573-4af1-990f-28e45e577c78 · inbound

Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation cites this paper.

Boosting Omni-Modal Language Models: Staged Post-Training with Visually Debiased Evaluation Step-Audio 2 Technical Report

Reference 24

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 10152838-7a19-4d13-83e7-092dc13b0b85 · inbound

Towards Fine-Grained Multi-Dimensional Speech Understanding: Data Pipeline, Benchmark, and Model cites this paper.

Towards Fine-Grained Multi-Dimensional Speech Understanding: Data Pipeline, Benchmark, and Model Step-Audio 2 Technical Report

Reference 26

Resolution: verified exact; arxiv_id, observed 2026-05-16T05:59:51.278263Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 93d478b3-29c8-4aa5-94b1-9b0d1317025c · inbound

A Survey of Advancing Audio Super-Resolution and Bandwidth Extension from Discriminative to Generative Models cites this paper.

A Survey of Advancing Audio Super-Resolution and Bandwidth Extension from Discriminative to Generative Models Step-Audio 2 Technical Report

Reference 64

Resolution: verified exact; local_arxiv, observed 2026-05-19T20:27:53.869969Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 1127c340-aff0-405a-a49c-8502257f8e5a · inbound

A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook cites this paper.

A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook Step-Audio 2 Technical Report

Reference 19

Resolution: verified exact; local_arxiv, observed 2026-05-21T07:39:48.989142Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a039000a-4505-47ea-9148-f65e54a56632 · inbound

DuplexSLA: A Full-Duplex Spoken Language Model with Synchronized Speech, Language, and Action cites this paper.

DuplexSLA: A Full-Duplex Spoken Language Model with Synchronized Speech, Language, and Action Step-Audio 2 Technical Report

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-05-21T02:43:55.050221Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 0ca85e37-9c8e-4e52-9c37-61d74f64ebdd · inbound

DuplexSLA: A Full-Duplex Spoken Language Model with Synchronized Speech, Language, and Action cites this paper.

DuplexSLA: A Full-Duplex Spoken Language Model with Synchronized Speech, Language, and Action Step-Audio 2 Technical Report

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-06-30T17:34:57.386787Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 5525e5c1-94a7-4295-b86b-3f417b2d5860 · inbound

A Survey of Audio Reasoning in Multimodal Foundation Models cites this paper.

A Survey of Audio Reasoning in Multimodal Foundation Models Step-Audio 2 Technical Report

Reference 61

Resolution: verified exact; local_arxiv, observed 2026-05-21T02:09:24.244009Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation ae35f11f-b45b-4570-a385-d37c6d821ce6 · inbound

StepAudio 2.5 Technical Report cites this paper.

StepAudio 2.5 Technical Report Step-Audio 2 Technical Report

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-05-25T02:55:16.487852Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation b7af688e-e9ac-4562-bcd4-937657082638 · inbound

EchoDistill:Alignment Noisy-to-Clean Self-Distillation for Robust Audio LLMs cites this paper.

EchoDistill:Alignment Noisy-to-Clean Self-Distillation for Robust Audio LLMs Step-Audio 2 Technical Report

Reference 50

Resolution: verified exact; local_arxiv, observed 2026-07-01T13:45:45.544097Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 343a53d0-a2e4-4ef5-874a-9ded9fad004a · inbound

Escape the Language Prior: Mitigating Late-Stage Modality Collapse in Audio Reasoning via Modality-Aware Policy Optimization cites this paper.

Escape the Language Prior: Mitigating Late-Stage Modality Collapse in Audio Reasoning via Modality-Aware Policy Optimization Step-Audio 2 Technical Report

Reference 24

Resolution: verified exact; local_arxiv, observed 2026-06-29T17:53:47.582917Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 74408ede-d0f8-4bbe-b9df-84d9d21dea95 · inbound

Audio-Mind: An Auditable Agentic Framework for Audio Understanding cites this paper.

Audio-Mind: An Auditable Agentic Framework for Audio Understanding Step-Audio 2 Technical Report

Reference 43

Resolution: verified exact; local_arxiv, observed 2026-06-29T10:13:17.937590Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 36336f64-05ab-4f78-97d4-f3975939abe4 · inbound

LaSR: Context-Aware Speech Recognition via Latent Reasoning cites this paper.

LaSR: Context-Aware Speech Recognition via Latent Reasoning Step-Audio 2 Technical Report

Reference 6

Resolution: metadata mismatch; local_arxiv, observed 2026-06-28T19:22:35.008192Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation ac64d043-d227-4e3f-ab87-0716ee761f16 · inbound

PolySpeech-100: A Large-Scale Benchmark for Speech Understanding Across 100+ Languages and Dialects cites this paper.

PolySpeech-100: A Large-Scale Benchmark for Speech Understanding Across 100+ Languages and Dialects Step-Audio 2 Technical Report

Reference 87

Resolution: verified exact; local_arxiv, observed 2026-06-28T17:52:27.042051Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 5961811b-46f4-4b63-9ac2-a0435eb4a629 · inbound

SpeechEditBench: A Bilingual Multi-Attribute Benchmark for Instruction-Guided Speech Editing cites this paper.

SpeechEditBench: A Bilingual Multi-Attribute Benchmark for Instruction-Guided Speech Editing Step-Audio 2 Technical Report

Reference 33

Resolution: metadata mismatch; local_arxiv, observed 2026-07-02T01:06:23.888556Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 12656078-064c-4202-892c-997079c380c7 · inbound

Audio Interaction Model cites this paper.

Audio Interaction Model Step-Audio 2 Technical Report

Reference 35

Resolution: verified exact; local_arxiv, observed 2026-07-02T10:46:52.396042Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation a2bd94ba-35b4-49bc-96d1-f7a23401665f · inbound

Beyond Semantic Dominance: Cognitive Affective Reasoning and Empathetic Response Alignment in Audio Language Models cites this paper.

Beyond Semantic Dominance: Cognitive Affective Reasoning and Empathetic Response Alignment in Audio Language Models Step-Audio 2 Technical Report

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-07-02T19:47:19.981292Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation b000e237-ce7b-4ff3-b2e9-9ec524cd0239 · inbound

Liberating LLM Capabilities in Full-Duplex Speech Models cites this paper.

Liberating LLM Capabilities in Full-Duplex Speech Models Step-Audio 2 Technical Report

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-07-01T00:15:08.998675Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4d14ecf7-34db-425e-aedd-406c859e58f7 · inbound

Multi-Faceted Interactivity Alignment in Full-Duplex Speech Models cites this paper.

Multi-Faceted Interactivity Alignment in Full-Duplex Speech Models Step-Audio 2 Technical Report

Reference 46

Resolution: verified exact; local_arxiv, observed 2026-06-27T13:20:57.416828Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 01340cbe-78c9-40a8-805f-1bef8fe4f259 · inbound

RAIL: Rethinking Auditory Intelligence in Large Audio-Language Models with a CHC-Grounded Benchmark cites this paper.

RAIL: Rethinking Auditory Intelligence in Large Audio-Language Models with a CHC-Grounded Benchmark Step-Audio 2 Technical Report

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-07-03T07:17:43.955861Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 60a0a846-d2e4-4ed9-8bfc-076ed6157a1c · inbound

Overcoming State Inertia in Full-Duplex Spoken Language Models via Activation Steering cites this paper.

Overcoming State Inertia in Full-Duplex Spoken Language Models via Activation Steering Step-Audio 2 Technical Report

Reference 42

Resolution: verified exact; local_arxiv, observed 2026-07-03T05:07:39.490323Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 4cbb51d7-5f78-4544-9922-3e5a57bf623e · inbound

Steering Where to Listen: Instruction-Based Activation Steering Redirects Temporal Attention in Large Audio-Language Models cites this paper.

Steering Where to Listen: Instruction-Based Activation Steering Redirects Temporal Attention in Large Audio-Language Models Step-Audio 2 Technical Report

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-07-03T07:57:44.795540Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation f48ebd15-b528-4404-8e8f-d9616782a1d4 · inbound

AOR-Bench: Do Large Audio Language Models Over-Refuse Pseudo-Harmful Queries? cites this paper.

AOR-Bench: Do Large Audio Language Models Over-Refuse Pseudo-Harmful Queries? Step-Audio 2 Technical Report

Reference 48

Resolution: metadata mismatch; local_arxiv, observed 2026-07-04T07:29:39.435997Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 2a929381-93cc-4823-9650-90f475d901b4 · inbound

MSU-Bench: Towards Speaker-Centric Understanding in Conversational Multi-Speaker Scenarios cites this paper.

MSU-Bench: Towards Speaker-Centric Understanding in Conversational Multi-Speaker Scenarios Step-Audio 2 Technical Report

Reference 32

Resolution: verified exact; local_arxiv, observed 2026-07-04T11:49:50.769665Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 9a40e8dc-484e-48cd-9fe0-820f4af639c8 · inbound

RedVox: Safety and Fairness Gaps in Speech Models Across Languages cites this paper.

RedVox: Safety and Fairness Gaps in Speech Models Across Languages Step-Audio 2 Technical Report

Reference 164

Resolution: verified exact; local_arxiv, observed 2026-07-04T13:59:52.900270Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 36bcd44d-52fb-4ae2-a3c7-da0abf707b31 · inbound

ALM2Vec: Learning Audio Embeddings for Universal Audio Retrieval with Large Audio-Language Models cites this paper.

ALM2Vec: Learning Audio Embeddings for Universal Audio Retrieval with Large Audio-Language Models Step-Audio 2 Technical Report

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-07-01T09:05:37.199775Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 1147bd70-7f4b-4557-a576-e455ba885218 · inbound

Preserving Speech-to-Text LLM Capabilities in Speech-to-Speech Generation cites this paper.

Preserving Speech-to-Text LLM Capabilities in Speech-to-Speech Generation Step-Audio 2 Technical Report

Reference 30

Resolution: verified exact; local_arxiv, observed 2026-07-01T13:05:45.819741Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation 0eabac8e-4d8b-4533-9cd6-a8a2bd50c0c7 · inbound

Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving cites this paper.

Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving Step-Audio 2 Technical Report

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-07-03T15:18:32.611833Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.

Observation d77b1801-15f9-4efb-884c-a642d666c852 · inbound

Unlocking Speech-Text Compositional Powers: Instruction-Following Speech Language Models without Instruction Tuning cites this paper.

Unlocking Speech-Text Compositional Powers: Instruction-Following Speech Language Models without Instruction Tuning Step-Audio 2 Technical Report

Reference 30

Resolution: verified exact; local_arxiv, observed 2026-07-03T14:38:28.609886Z

Source-reported events for the cited work

Unavailable: named source frontier unavailable.