The 2027 Land-Doig MIP Competition. Explainability and Small Infeasible Subsystems. January 22, 2027 (AoE) - Registration closes
www.mixedinteger.org/2027/competi...
www.mixedinteger.org/2027/competi...
October 30, 2026 at 10:16 AM
The 2027 Land-Doig MIP Competition. Explainability and Small Infeasible Subsystems. January 22, 2027 (AoE) - Registration closes
www.mixedinteger.org/2027/competi...
www.mixedinteger.org/2027/competi...
A multimodal vision language foundation model, EVLF-FM, integrates extensive clinical data and advanced training techniques, achieving superior diagnostic accuracy and explainability across diverse imaging modalities.
A clinical multimodal vision language foundation model with fine-grain explainability - Nature Medicine
A multimodal vision language foundation model, EVLF-FM, integrates extensive clinical data and advanced training techniques, achieving superior diagnostic accuracy and explainability across diverse imaging modalities, enhancing clinical decision-making through robust reasoning and interpretability.
dlvr.it
October 30, 2026 at 3:01 PM
A multimodal vision language foundation model, EVLF-FM, integrates extensive clinical data and advanced training techniques, achieving superior diagnostic accuracy and explainability across diverse imaging modalities.
Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study
Robbie Southam, Zhong Fan, Ke Li
#arXiv #cs.AI #cs.LG
Robbie Southam, Zhong Fan, Ke Li
#arXiv #cs.AI #cs.LG
Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study
As the effects of climate change become increasingly apparent, the demand for renewable energy sources such as wind power continues to grow. However, renewables' variable and intermittent nature presents significant challenges in their integration into energy grids. Artificial intelligence (AI) has…
arxiv.org
November 1, 2026 at 2:19 AM
💼 The revival of retro tech like Teletype and Decwriter highlights the tension between cost-effective AI adoption and the need for explainability, raising questions about the long-term viability of black box solutions.
November 2, 2026 at 12:36 PM
💼 The revival of retro tech like Teletype and Decwriter highlights the tension between cost-effective AI adoption and the need for explainability, raising questions about the long-term viability of black box solutions.
My own personal work looks at experts who use AI to create what-if simulations to help inform their decisions. I am looking at explainability and interpretability exactly because I know experts have cognitive biases that AI can trigger. I know because I’ve read cognitive science and HCI literature.
October 28, 2026 at 11:19 PM
My own personal work looks at experts who use AI to create what-if simulations to help inform their decisions. I am looking at explainability and interpretability exactly because I know experts have cognitive biases that AI can trigger. I know because I’ve read cognitive science and HCI literature.
Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study
Robbie Southam, Zhong Fan, Ke Li
#arXiv #cs.AI #cs.LG
Robbie Southam, Zhong Fan, Ke Li
#arXiv #cs.AI #cs.LG
Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study
As the effects of climate change become increasingly apparent, the demand for renewable energy sources such as wind power continues to grow. However, renewables' variable and intermittent nature presents significant challenges in their integration into energy grids. Artificial intelligence (AI) has…
arxiv.org
October 31, 2026 at 12:52 AM
78% of organizations claim full trust in AI. Only 40% invest in governance, explainability, and safeguards. Only 2% rank a governance framework as a top priority. Source: IDC. #Valkyree
October 30, 2026 at 2:02 PM
78% of organizations claim full trust in AI. Only 40% invest in governance, explainability, and safeguards. Only 2% rank a governance framework as a top priority. Source: IDC. #Valkyree
Robbie Southam, Zhong Fan, Ke Li: Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study https://arxiv.org/abs/2610.40330 https://arxiv.org/pdf/2610.40330 https://arxiv.org/html/2610.40330
October 30, 2026 at 6:46 AM
Robbie Southam, Zhong Fan, Ke Li: Multistep and Multivariate Time Series Explainability with GradientSHAP: A Wind Power Case Study https://arxiv.org/abs/2610.40330 https://arxiv.org/pdf/2610.40330 https://arxiv.org/html/2610.40330
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
October 29, 2026 at 8:34 AM
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
Does Federated Training Degrade Explainability? A Comparative Analysis of Grad-CAM Faithfulness in Centralized vs. Federated Deepfake Detection
Muhammad Uzair Shahid, Muhammad Zabil Mehboob
#arXiv #cs.AI #cs.CR #cs.CV #cs.LG
Muhammad Uzair Shahid, Muhammad Zabil Mehboob
#arXiv #cs.AI #cs.CR #cs.CV #cs.LG
Does Federated Training Degrade Explainability? A Comparative Analysis of Grad-CAM Faithfulness in Centralized vs. Federated Deepfake Detection
Automated detection has become the essential safeguard against realistic synthetic media, but deploying it at scale requires training on sensitive data from multiple organizations, putting privacy and the transparency demanded by forensic review in direct conflict. Federated learning resolves the p…
arxiv.org
October 31, 2026 at 4:23 PM
@gui.do They, like me, and anyone else who has to work with the results of LLM spewed apps flung out by every high schooler who can't pass grade 10 math, are being overwhelmed by nonsense, right from the submission of applications.
This is a requirement for the TEA: transparency, explainability […]
This is a requirement for the TEA: transparency, explainability […]
Original post on cosocial.ca
cosocial.ca
October 29, 2026 at 2:04 AM
@gui.do They, like me, and anyone else who has to work with the results of LLM spewed apps flung out by every high schooler who can't pass grade 10 math, are being overwhelmed by nonsense, right from the submission of applications.
This is a requirement for the TEA: transparency, explainability […]
This is a requirement for the TEA: transparency, explainability […]
Eduard Barbu, Rodion Krjut\v{s}kov, Anneli Uusk\"ula: Evaluating Conversational Explainability in Healthcare: Comparing the Explainability Assistant and a Traditional XAI Dashboard https://arxiv.org/abs/2610.39634 https://arxiv.org/pdf/2610.39634 https://arxiv.org/html/2610.39634
October 30, 2026 at 6:42 AM
Eduard Barbu, Rodion Krjut\v{s}kov, Anneli Uusk\"ula: Evaluating Conversational Explainability in Healthcare: Comparing the Explainability Assistant and a Traditional XAI Dashboard https://arxiv.org/abs/2610.39634 https://arxiv.org/pdf/2610.39634 https://arxiv.org/html/2610.39634
Gol Mohammad Dorrazehi from Cambridge combines directed evolution and deep mutational scanning to retain negative results when evolving new proteins. This helps with explainability and iteration of libraries
November 2, 2026 at 12:09 PM
Gol Mohammad Dorrazehi from Cambridge combines directed evolution and deep mutational scanning to retain negative results when evolving new proteins. This helps with explainability and iteration of libraries
What stands out?
+ End-to-end curiosity: Understand the data architecture;
+ Explainability: Walk through your decision logic;
+ Problem context: Frame projects around real business trade-offs;
+ Foundational resilience: Adapt as tools evolve.
+ End-to-end curiosity: Understand the data architecture;
+ Explainability: Walk through your decision logic;
+ Problem context: Frame projects around real business trade-offs;
+ Foundational resilience: Adapt as tools evolve.
November 2, 2026 at 2:30 PM
What stands out?
+ End-to-end curiosity: Understand the data architecture;
+ Explainability: Walk through your decision logic;
+ Problem context: Frame projects around real business trade-offs;
+ Foundational resilience: Adapt as tools evolve.
+ End-to-end curiosity: Understand the data architecture;
+ Explainability: Walk through your decision logic;
+ Problem context: Frame projects around real business trade-offs;
+ Foundational resilience: Adapt as tools evolve.
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
October 29, 2026 at 8:34 AM
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
October 29, 2026 at 8:34 AM
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
October 29, 2026 at 8:34 AM
When healthcare AI takes action, can you reconstruct what happened? Audit trails should capture relevant evidence, permissions, tool calls, outcomes and human interventions. Explainability helps interpret outputs. Auditability helps verify events and accountability.
Term of the day: Model Explainability
Model explainability is the ability to describe how an artificial intelligence (AI) or machine learning (ML) system produces its outputs, in a way that makes sense to...
Full definition: compliancebureau.org/glossary/model-explainability
#compliance #GRC
Model explainability is the ability to describe how an artificial intelligence (AI) or machine learning (ML) system produces its outputs, in a way that makes sense to...
Full definition: compliancebureau.org/glossary/model-explainability
#compliance #GRC
October 28, 2026 at 3:08 PM
Term of the day: Model Explainability
Model explainability is the ability to describe how an artificial intelligence (AI) or machine learning (ML) system produces its outputs, in a way that makes sense to...
Full definition: compliancebureau.org/glossary/model-explainability
#compliance #GRC
Model explainability is the ability to describe how an artificial intelligence (AI) or machine learning (ML) system produces its outputs, in a way that makes sense to...
Full definition: compliancebureau.org/glossary/model-explainability
#compliance #GRC
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
AI assistants are increasingly embedded in computational notebooks, scientific workflows, software engineering tools, and learning systems, yet human-AI interactions are often treated as transient prompt-response exchanges rather than durable data products, hindering reproducibility, explainability…
arxiv.org
October 31, 2026 at 9:27 AM
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
AI assistants are increasingly embedded in computational notebooks, scientific workflows, software engineering tools, and learning systems, yet human-AI interactions are often treated as transient prompt-response exchanges rather than durable data products, hindering reproducibility, explainability…
arxiv.org
October 31, 2026 at 7:04 AM
The paper presents a 24-hour ambulatory screening framework for subcutaneous and extravascular ICDs and introduces AmpRatioNet, a 1-D CNN operating on ECG signals.
Source: arXiv cs.LG
Source: arXiv cs.LG
Deep learning for non-transvenous ICD screening: tailored neural network architectures, extended experiments, and explainability
Abstract page for arXiv paper 2610.37993: Deep learning for non-transvenous ICD screening: tailored neural network architectures, extended experiments, and explainability
arxiv.org
October 30, 2026 at 7:36 AM
The paper presents a 24-hour ambulatory screening framework for subcutaneous and extravascular ICDs and introduces AmpRatioNet, a 1-D CNN operating on ECG signals.
Source: arXiv cs.LG
Source: arXiv cs.LG
1/2
ABR explainability = less guesswork.
Quick fields to surface in your logs/analytics:
• last_decision (why a rendition was chosen)
• measured_buffer_seconds at decision time
• bandwidth_sample + timestamp
• device_class hint (memory/cpu)
ABR explainability = less guesswork.
Quick fields to surface in your logs/analytics:
• last_decision (why a rendition was chosen)
• measured_buffer_seconds at decision time
• bandwidth_sample + timestamp
• device_class hint (memory/cpu)
October 30, 2026 at 6:57 AM
1/2
ABR explainability = less guesswork.
Quick fields to surface in your logs/analytics:
• last_decision (why a rendition was chosen)
• measured_buffer_seconds at decision time
• bandwidth_sample + timestamp
• device_class hint (memory/cpu)
ABR explainability = less guesswork.
Quick fields to surface in your logs/analytics:
• last_decision (why a rendition was chosen)
• measured_buffer_seconds at decision time
• bandwidth_sample + timestamp
• device_class hint (memory/cpu)
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
AI assistants are increasingly embedded in computational notebooks, scientific workflows, software engineering tools, and learning systems, yet human-AI interactions are often treated as transient prompt-response exchanges rather than durable data products, hindering reproducibility, explainability…
arxiv.org
October 30, 2026 at 7:11 AM
1/2
Attach explainability metadata to every model reply so support and audits move faster.
Include:
- prompt_hash + prompt_version
- model_version + temperature
- top_sources (ids) + retrieval_confidence
- confidence_hint (low|medium|high) + suggested_next_action (ask_clarify|fallback)
Attach explainability metadata to every model reply so support and audits move faster.
Include:
- prompt_hash + prompt_version
- model_version + temperature
- top_sources (ids) + retrieval_confidence
- confidence_hint (low|medium|high) + suggested_next_action (ask_clarify|fallback)
October 29, 2026 at 8:38 PM
1/2
Attach explainability metadata to every model reply so support and audits move faster.
Include:
- prompt_hash + prompt_version
- model_version + temperature
- top_sources (ids) + retrieval_confidence
- confidence_hint (low|medium|high) + suggested_next_action (ask_clarify|fallback)
Attach explainability metadata to every model reply so support and audits move faster.
Include:
- prompt_hash + prompt_version
- model_version + temperature
- top_sources (ids) + retrieval_confidence
- confidence_hint (low|medium|high) + suggested_next_action (ask_clarify|fallback)
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Michela Taufer et al.
#arXiv #cs.AI #cs.HC
Human--AI Interactions as First-Class Data: A Formal Interaction Model and Runtime for Explainable AI Collaboration
AI assistants are increasingly embedded in computational notebooks, scientific workflows, software engineering tools, and learning systems, yet human-AI interactions are often treated as transient prompt-response exchanges rather than durable data products, hindering reproducibility, explainability…
arxiv.org
October 30, 2026 at 9:40 AM