Paper Citation Record · LEDGER

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions

As of 20 August 2026, this Paper Citation Record lists 37 of 37 outbound references and 1 inbound Pith citation observation for arXiv:2506.09556.

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

pith.paper-citation-record.v1
2506.09556 v2

Coverage vector

measured 37 of 37 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T04:49:09.335889Z

measured 38 of 38 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-20T06:33:59.587034+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T04:49:06.392265Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-08-07T04:49:09.594801Z

Reference resolution

37 of 37 outbound references displayed

verified exact1
verified fuzzy28
unresolved7
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation d16c6d11-83fe-4131-bab5-2ec70e60726d · outbound

This paper cites Categorical Emo- tion Recognition.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Categorical Emo- tion Recognition

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:15.822861Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 9badd1b9-a52a-469a-9d80-46af3ab08019 · outbound

This paper cites MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions

Reference 2

Resolution: metadata mismatch; local_arxiv, observed 2026-08-07T04:49:09.666502Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 12d57c54-07d0-4e7d-ae7f-fe4e76ad3cb5 · outbound

This paper cites Overview The novelty of our approach stems from three key directions: 1) architecture, 2) data utilization and 3) training recipe.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Overview The novelty of our approach stems from three key directions: 1) architecture, 2) data utilization and 3) training recipe

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:15.724599Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 0494fa24-9d83-4d75-81d0-814cf1cf5346 · outbound

This paper cites The Stage 3 meta-classifier uses batch size 128 and learning rate 1e-3 without M.MixUp.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions The Stage 3 meta-classifier uses batch size 128 and learning rate 1e-3 without M.MixUp

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:15.598901Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation dd6468ea-74ef-4c5e-8316-e55064c069c6 · outbound

This paper cites an unresolved cited work.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Unresolved cited work

Reference 5

Resolution: unresolved; raw_fallback, observed 2026-08-07T04:49:15.480102Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 6a057bc7-41f9-428b-939f-0407ed8b9e8c · outbound

This paper cites an unresolved cited work.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Unresolved cited work

Reference 6

Resolution: unresolved; raw_fallback, observed 2026-08-07T04:49:15.359673Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 69592d8d-9fb1-4522-9740-adcab9a7dd85 · outbound

This paper cites an unresolved cited work.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Unresolved cited work

Reference 7

Resolution: unresolved; raw_fallback, observed 2026-08-07T04:49:15.161502Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 0dad5a84-0db1-4a82-a23b-e883d17b77a9 · outbound

This paper cites Emotion recognition in human- computer interaction,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Emotion recognition in human- computer interaction,

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:14.924694Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 36caba0a-ec5c-464b-a2d8-a4072a2a11fc · outbound

This paper cites Make patient consultation warmer: A clinical application for speech emotion recognition,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Make patient consultation warmer: A clinical application for speech emotion recognition,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:14.793854Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation ddd14370-4c88-4304-98eb-0643ee3f4472 · outbound

This paper cites Integration of driver be- havior into emotion recognition systems: A preliminary study on steering wheel and vehicle acceleration,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Integration of driver be- havior into emotion recognition systems: A preliminary study on steering wheel and vehicle acceleration,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:14.628524Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e944c8ae-e12a-4a23-bc6b-7b3964a3cdb8 · outbound

This paper cites Call redistribution for a call center based on speech emotion recognition,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Call redistribution for a call center based on speech emotion recognition,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:14.462141Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 26a9d0b5-8444-4c67-9620-331f874df050 · outbound

This paper cites emotion2vec: Self-supervised pre-training for speech emotion representation,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions emotion2vec: Self-supervised pre-training for speech emotion representation,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:14.265007Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 1e4348d4-2ca5-4039-aea8-8ffe4ab2c932 · outbound

This paper cites Adieu features? end-to-end speech emotion recognition using a deep convolutional recurrent network,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Adieu features? end-to-end speech emotion recognition using a deep convolutional recurrent network,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:14.034592Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation d1dd1b22-bb67-460b-9f0f-844433cf70d7 · outbound

This paper cites The interspeech 2025 challenge on speech emotion recognition in naturalistic conditions,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions The interspeech 2025 challenge on speech emotion recognition in naturalistic conditions,

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:13.844521Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 3e8f70a8-a254-4928-86a2-3f8df4e160f9 · outbound

This paper cites Building naturalistic emotionally bal- anced speech corpus by retrieving emotional speech from existing podcast recordings,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Building naturalistic emotionally bal- anced speech corpus by retrieving emotional speech from existing podcast recordings,

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:13.649321Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 462c9e3d-d649-46fd-bb64-ddcb91d79659 · outbound

This paper cites Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:13.366793Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 0a85acd4-f961-4969-a404-5ef36734388f · outbound

This paper cites Deep hier- archical fusion with application in sentiment analysis,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Deep hier- archical fusion with application in sentiment analysis,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:13.125839Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation c1d370ad-ad16-49ac-9a81-8fccc442f7cd · outbound

This paper cites Iemocap: Interactive emotional dyadic motion capture database,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Iemocap: Interactive emotional dyadic motion capture database,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:12.902567Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 8a41fd6d-e4bf-403f-970d-53cc5e70328f · outbound

This paper cites Multimodal language analysis in the wild: Cmu-mosei dataset and interpretable dynamic fusion graph,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Multimodal language analysis in the wild: Cmu-mosei dataset and interpretable dynamic fusion graph,

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:12.558675Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 70fe6991-914a-48e8-9c93-61251e71b6df · outbound

This paper cites A survey of speech emotion recognition in natural environment,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions A survey of speech emotion recognition in natural environment,

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:12.291713Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation b60dca6a-993c-4f54-b86f-e1dc96de0065 · outbound

This paper cites WavLLM: Towards robust and adaptive speech large language model,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions WavLLM: Towards robust and adaptive speech large language model,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:12.041077Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 4d577b80-37f9-45ed-aba9-e89ed2092976 · outbound

This paper cites Soft-target training with ambiguous emotional ut- terances for dnn-based speech emotion classification,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Soft-target training with ambiguous emotional ut- terances for dnn-based speech emotion classification,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:11.764424Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 815d5e63-b387-41a7-b82a-e473abebfc6b · outbound

This paper cites Manifold mixup: Better represen- tations by interpolating hidden states,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Manifold mixup: Better represen- tations by interpolating hidden states,

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:11.484741Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation b4770892-23f7-424b-aa38-1c9c8141cd32 · outbound

This paper cites Powmix: A versa- tile regularizer for multimodal sentiment analysis,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Powmix: A versa- tile regularizer for multimodal sentiment analysis,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:11.266365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 4d07d30a-53a2-4414-b0d1-c55a0835069b · outbound

This paper cites Speech emotion recognition with multi-task learning,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Speech emotion recognition with multi-task learning,

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:11.110759Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation a0b28d41-bf57-4949-aeb7-86e82cddf8fe · outbound

This paper cites 1st Place Solution to Odyssey Emotion Recognition Challenge Task1: Tackling Class Imbalance Problem.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions 1st Place Solution to Odyssey Emotion Recognition Challenge Task1: Tackling Class Imbalance Problem

Reference 26

Resolution: verified exact; local_arxiv, observed 2026-08-07T04:49:09.548351Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 07c27be4-6c06-4774-b74d-2b0d0210aacd · outbound

This paper cites Meta-classifiers easily im- prove commercial sentiment detection tools,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Meta-classifiers easily im- prove commercial sentiment detection tools,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:10.972515Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation d4f1aed8-ee61-4753-9e18-001f4ba3cb8e · outbound

This paper cites Attention is all you need,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Attention is all you need,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:10.786076Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 719d48d6-3b68-4008-9942-34204805d496 · outbound

This paper cites Wavlm: Large-scale self-supervised pre-training for full stack speech processing,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Wavlm: Large-scale self-supervised pre-training for full stack speech processing,

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:10.600445Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation d896ca3b-cecd-4987-a3a0-3f472e2848c9 · outbound

This paper cites Hubert: Self-supervised speech represen- tation learning by masked prediction of hidden units,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Hubert: Self-supervised speech represen- tation learning by masked prediction of hidden units,

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-07T04:49:08.809717Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7a0a019-5306-4013-80b2-77b90dbf5bfe · outbound

This paper cites Robust speech recognition via large-scale weak su- pervision,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Robust speech recognition via large-scale weak su- pervision,

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-07T04:49:08.900960Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d6105cb-5681-4626-9ecb-d3e791be2d64 · outbound

This paper cites Roberta: A robustly optimized bert pretraining approach,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Roberta: A robustly optimized bert pretraining approach,

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-07T04:49:08.985412Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f6fa6fb4-aab2-4edd-b0c2-793f97481112 · outbound

This paper cites Smarter, better, faster, longer: A modern bidirectional encoder for fast, memory efficient, and long context finetuning and inference,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Smarter, better, faster, longer: A modern bidirectional encoder for fast, memory efficient, and long context finetuning and inference,

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:10.433802Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation cf9926c2-6b78-4f5a-8195-5296dfea9d97 · outbound

This paper cites Long short-term memory,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Long short-term memory,

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-07T04:49:09.118064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 344cae71-43c8-46cd-8ee7-93313d28eced · outbound

This paper cites Vicinal risk minimization,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Vicinal risk minimization,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:10.253030Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 8add645e-4b2c-443b-b366-0b70293a58ce · outbound

This paper cites Decoupled weight decay regulariza- tion,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Decoupled weight decay regulariza- tion,

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:10.084078Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation b42fd3f9-8335-4f7a-8ef7-22e8bcfa5736 · outbound

This paper cites Py- torch: An imperative style, high-performance deep learning li- brary,.

MEDUSA: A Multimodal Deep Fusion Multi-Stage Training Framework for Speech Emotion Recognition in Naturalistic Conditions Py- torch: An imperative style, high-performance deep learning li- brary,

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T04:49:09.907373Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Pith citing papers

Observation 9badd1b9-a52a-469a-9d80-46af3ab08019 · inbound