MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection

MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection | Signal Canvas | ScienceToStartup

Page Freshness

Signal Canvas proof surface

Canonical route: /signal-canvas/mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio

stale

Proof freshness: stale
Proof status: unverified
Display score: 7/10
Last proof check: 2026-03-30
Score updated: 2026-04-02
Score fresh until: 2026-05-02
References: 57
Source count: 3
Coverage: 50%

This page is showing the last landed evidence receipt and score bundle because the latest proof data is outside the freshness window.

Agent Handoff

Canonical ID mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio | Route /signal-canvas/mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio

REST example

curl https://sciencetostartup.com/api/v1/agent-handoff/signal-canvas/mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio

MCP example

{
  "tool": "search_signal_canvas",
  "arguments": {
    "mode": "paper",
    "paper_ref": "mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio",
    "query_text": "Summarize MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection"
  }
}

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection",
  "normalized_query": "2603.26064",
  "route": "/signal-canvas/mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio",
  "paper_ref": "mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}

Evidence Receipt

Route status: building

Claims: 12

References: 57

Proof: Verification pending

Freshness state: computing

Source paper: MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection

PDF: https://arxiv.org/pdf/2603.26064v1

Source count: 3

Coverage: 50%

Last proof check: 2026-03-30T22:25:04.723Z

Signal Canvas receipt window

Watch and verify: MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection

/buildability/mudd-a-multimodal-deception-detection-dataset-and-gsr-guided-progressive-distillation-for-non-contact-deception-detectio

Watchwatch

Subject: MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection

Verdict

Watch

Preparing verified analysis

GitHub Code Pulse

No public code linked for this paper yet.

Claim map

Strong 12Mixed 0Weak 0

Evidencepartial
In addition to video, audio, and GSR, MuDD also provides Photoplethysmography, heart rate, and personality traits, supporting broader scientific studies of deception.
Implicationpartial
The abstract and dataset description explicitly state the size and modalities of the MuDD dataset.
Verificationpartial
partial
Evidencepartial
Based on this dataset, we propose GSR-guided Progressive Distillation (GPD), a cross-modal distillation framework for non-contact deception detection.
Implicationpartial
The abstract and introduction clearly define GPD and its purpose.
Verificationpartial
partial
Evidencepartial
The core innovation of GPD is the integration of progressive feature-level and digit-level distillation with dynamic routing, which allows the model to adaptively determine how teacher knowledge should be transferred during training, leading to more stable cross-modal knowledge transfer.
Implicationpartial
The abstract and method description detail the core components of GPD.
Verificationpartial
partial
Evidencepartial
Extensive experiments and visualizations show that GPD outperforms existing methods and achieves state-of-the-art performance on both deception detection and concealed-digit identification.
Implicationpartial
The abstract states that GPD achieves state-of-the-art performance, and the results section indicates comparisons against baselines.
Verificationpartial
partial
Evidencepartial
Non-contact automatic deception detection remains challenging because visual and auditory deception cues often lack stable cross-subject patterns. In contrast, galvanic skin response (GSR) provides more reliable physiological cues and has been widely used in contact-based deception detection.
Implicationpartial
The abstract highlights the advantage of GSR over non-contact modalities for deception detection.
Verificationpartial
partial
Evidencepartial
GPD consists of four key components: a teacher network, a student network, a progressive knowledge distillation module, and a digit-level evidence aggregation module. The teacher network takes GSR as input, while the student network takes visual and audio features as input.
Implicationpartial
The description of GPD's components is provided in the text.
Verificationpartial
partial
Evidencepartial
GPD distills two types of teacher knowledge: feature-level physiological representations and digit-level concealed-item evidence.
Implicationpartial
The abstract and method description explicitly mention the two types of knowledge being distilled.
Verificationpartial
partial
Evidencepartial
In addition to video, audio, and GSR, MuDD also provides Photoplethysmography, heart rate, and personality traits, supporting broader scientific studies of deception.
Implicationpartial
The abstract explicitly states the size and content of the MuDD dataset, and this is further detailed in the 'MuDD Dataset' section.
Verificationpartial
partial
Evidencepartial
Based on this dataset, we propose GSR-guided Progressive Distillation (GPD), a cross-modal distillation framework for non-contact deception detection.
Implicationpartial
The abstract and introduction clearly define GPD as a method for non-contact deception detection using GSR guidance.
Verificationpartial
partial
Evidencepartial
The core innovation of GPD is the integration of progressive feature-level and digit-level distillation with dynamic routing, which allows the model to adaptively determine how teacher knowledge should be transferred during training, leading to more stable cross-modal knowledge transfer.
Implicationpartial
The abstract and description of GPD highlight these key components and their adaptive nature.
Verificationpartial
partial
Evidencepartial
Extensive experiments and visualizations show that GPD outperforms existing methods and achieves state-of-the-art performance on both deception detection and concealed-digit identification.
Implicationpartial
The abstract states that GPD achieves state-of-the-art performance, and the 'Main Results' section indicates comparisons against existing methods.
Verificationpartial
partial
Evidencepartial
Extensive experiments and visualizations show that GPD outperforms existing methods and achieves state-of-the-art performance on both deception detection and concealed-digit identification.
Implicationpartial
The abstract explicitly states that GPD outperforms existing methods, and the 'Main Results' section is dedicated to comparing it against baselines.
Verificationpartial
partial

Author intelligence and commercialization panels stay hidden until the proof receipt is verified, cites at least 3 references, includes at least 2 sources, and clears 50% coverage. The paper narrative and citation surfaces remain public while verification is pending.

MuDD: A Multimodal Deception Detection Dataset and GSR-Guided Progressive Distillation for Non-Contact Deception Detection

Use Signal Canvas as the narrative proof surface

Use this Signal Canvas via API or MCP

Signal Canvas proof surface