🤖 AI Research - Physiological Signal Processing

GeoRVQ: Decoder-aware geometry for residual-token prediction in physiological signals
Bo Cui, Yaowen Zhang
https://arxiv.org/abs/2609.27018 ↗

GeoRVQ is a novel approach to residual-token prediction in physiological signals. It leverages decoder-aware geometry to improve the accuracy of predictions.

Key Points:

Decoder-aware geometry: GeoRVQ uses a geometric approach to model the relationship between the input signal and the decoder. This allows for more accurate predictions.
Residual-token prediction: GeoRVQ is designed to predict the residual tokens in physiological signals, which can be used to improve the accuracy of signal processing.
Physiological signals: GeoRVQ is specifically designed for use with physiological signals, such as ECG and EEG.

🔗 Resources:

Image

🚀 AI Innovation Lab - Upcoming Events

Thought we were done travelling this year? Not quite
Next up on the @Fetch_ai Innovation Lab activation list:
@mhacks (Ann Arbor)
@LAHacks AI Hackathon (Los Angeles)
Fetch-A-Thon 3.0 (India)
@TheBusinessShow (London)
Hackathons, business events and a lot of

The Fetch.ai Innovation Lab is excited to announce its upcoming events, including hackathons and business events.

Key Points:

Upcoming events: The Fetch.ai Innovation Lab has a number of upcoming events, including hackathons and business events.
@mhacks: The @mhacks event will be held in Ann Arbor.
@LAHacks AI Hackathon: The @LAHacks AI Hackathon will be held in Los Angeles.
Fetch-A-Thon 3.0: Fetch-A-Thon 3.0 will be held in India.
@TheBusinessShow: @TheBusinessShow will be held in London.

🔗 Resources:

Image

🤖 AI - Personal AI

One ASI:One account gives you a Personal AI automatically.
No, you don’t need a developer diploma.
Or a separate Agent configuration.
(1) Go to https://asi1.ai ↗.
(2) Sign up and personalize.
(3) Start using it.
It begins learning your style, interests and preferences as

One ASI is a personal AI that can be accessed through a single account. It does not require a developer diploma or separate agent configuration.

Key Points:

Personal AI: One ASI is a personal AI that can be accessed through a single account.
No developer diploma required: One ASI does not require a developer diploma or separate agent configuration.
Personalization: One ASI can be personalized to learn the user's style, interests, and preferences.

🔗 Resources:

Image

🤖 AI - Access Control for Tool-Using LLM Agents

Progressive Skill Discovery as Access Control for Tool-Using LLM Agents: Structural Governance through Role-Scoped Capability Delivery
Michael Stettler, Benjamin Girardet, Jonas Canton, Nicolas Corod
https://arxiv.org/abs/2609.28693 ↗

Progressive Skill Discovery is a novel approach to access control for tool-using LLM agents. It uses structural governance to deliver role-scoped capability delivery.

Key Points:

Progressive Skill Discovery: Progressive Skill Discovery is a novel approach to access control for tool-using LLM agents.
Structural governance: Progressive Skill Discovery uses structural governance to deliver role-scoped capability delivery.
Role-scoped capability delivery: Progressive Skill Discovery delivers role-scoped capability delivery to LLM agents.

🔗 Resources:

Image

🤖 AI - Dexterous Robot Manipulation

Dexterous Robot Manipulation from Human Demonstrations via Contact-Anchored Retargeting and Residual Policy Learning
Zihao Yang, Chengyuan Liu, Yu Zhou, Runze Lv, Tianyu Cui, Sheng Yi, Haohua Zhu, Irvine Lu, ...
https://arxiv.org/abs/2609.24093 ↗

Dexterous Robot Manipulation is a novel approach to dexterous robot manipulation. It uses contact-anchored retargeting and residual policy learning to improve the accuracy of manipulation.

Key Points:

Dexterous Robot Manipulation: Dexterous Robot Manipulation is a novel approach to dexterous robot manipulation.
Contact-anchored retargeting: Dexterous Robot Manipulation uses contact-anchored retargeting to improve the accuracy of manipulation.
Residual policy learning: Dexterous Robot Manipulation uses residual policy learning to improve the accuracy of manipulation.

🔗 Resources:

Image

🎉 Congratulations to Sudipto Ghosh

Congratulations to Sudipto Ghosh (@ScientificGhosh) on being awarded the prestigious @IndiaDST #INSPIRE #PhDFellowship!
We are proud of his achievement and look forward to his continued contributions and success in the years ahead!

Sudipto Ghosh has been awarded the prestigious @IndiaDST #INSPIRE #PhDFellowship.

Key Points:

Congratulations to Sudipto Ghosh: Sudipto Ghosh has been awarded the prestigious @IndiaDST #INSPIRE #PhDFellowship.
@IndiaDST #INSPIRE #PhDFellowship: The @IndiaDST #INSPIRE #PhDFellowship is a prestigious award.
Proud of his achievement: The team is proud of Sudipto Ghosh's achievement.

🔗 Resources:

Image

🤖 AI - Resource-Efficient Distributed Recursive Gaussian Processes

Resource-Efficient Distributed Recursive Gaussian Processes
Josephine King, Ali Emre Balci, Raj Thilak Rajan
https://arxiv.org/abs/2609.26979 ↗

Resource-Efficient Distributed Recursive Gaussian Processes is a novel approach to resource-efficient distributed recursive Gaussian processes. It uses a novel approach to improve the accuracy of Gaussian processes.

Key Points:

Resource-Efficient Distributed Recursive Gaussian Processes: Resource-Efficient Distributed Recursive Gaussian Processes is a novel approach to resource-efficient distributed recursive Gaussian processes.
Novel approach: Resource-Efficient Distributed Recursive Gaussian Processes uses a novel approach to improve the accuracy of Gaussian processes.
Improved accuracy: Resource-Efficient Distributed Recursive Gaussian Processes improves the accuracy of Gaussian processes.

🔗 Resources:

Image

🤖 AI - Driving Epidemic Models with AI Agents

Driving Epidemic Models with AI Agents: the Epydemix Agent Framework
Nicolò Gozzi, Ciro Cattuto, Alessandro Vespignani
https://arxiv.org/abs/2609.28692 ↗

Driving Epidemic Models with AI Agents is a novel approach to driving epidemic models with AI agents. It uses the Epydemix Agent Framework to improve the accuracy of epidemic models.

Key Points:

Driving Epidemic Models with AI Agents: Driving Epidemic Models with AI Agents is a novel approach to driving epidemic models with AI agents.
Epydemix Agent Framework: Driving Epidemic Models with AI Agents uses the Epydemix Agent Framework to improve the accuracy of epidemic models.
Improved accuracy: Driving Epidemic Models with AI Agents improves the accuracy of epidemic models.

🔗 Resources:

Image

🤖 AI - When Does Touch Matter?

When Does Touch Matter? Charting the Vision-Interaction Gap in Cluttered Dexterous Grasping
Hao Jiang, Luis Dominguez, Daniel Seita
https://arxiv.org/abs/2609.24068 ↗

When Does Touch Matter? is a novel approach to charting the vision-interaction gap in cluttered dexterous grasping. It uses a novel approach to improve the accuracy of grasping.

Key Points:

When Does Touch Matter?: When Does Touch Matter? is a novel approach to charting the vision-interaction gap in cluttered dexterous grasping.
Novel approach: When Does Touch Matter? uses a novel approach to improve the accuracy of grasping.
Improved accuracy: When Does Touch Matter? improves the accuracy of grasping.

🔗 Resources:

Image

🚀 Environmental Hacks

The second hackathon of the Bharat Builds Tour, Environmental Hacks, encourages students across India to build practical solutions to environmental challenges in their communities
Choose from three tracks:

Air : Build solutions for air quality monitoring, pollution
Water : Build solutions for water conservation, management
Waste : Build solutions for waste management, recycling

The Bharat Builds Tour is excited to announce its second hackathon, Environmental Hacks.

Key Points:

Environmental Hacks: The Bharat Builds Tour is