Source Identification Is Not Fitness Testing: Measuring the Limits of Synthetic-Data Attribution
Paper • 2610.00417 • Published • 1 •
3
Joss Armstrong
jarmstrong001
·
AI & ML interests
Information limits in multi-agent coordination, task-relevant information for machine interfaces, O-RAN autonomous network management, calibration of LLM confidence.
Recent Activity
commentedon a paper about 8 hours ago
Source Identification Is Not Fitness Testing: Measuring the Limits of Synthetic-Data Attribution liked a Space about 11 hours ago
huggingface/paper-central upvoted a paper about 11 hours ago
Self-Consuming Generative Models Go MADOrganizations
None yet
commented a
paper about 8 hours ago
liked a
Space about 11 hours ago
upvoted 3
papers about 11 hours ago
liked 3
datasets about 11 hours ago
submitted a
paper to Daily Papers about 11 hours ago
authored 6
papers about 12 hours ago
Source-Side Sufficiency for the Information Bottleneck: Exact Reduction and Finite-Block Equivalence
Paper • 2604.26744 • Published
Implicit Evaluation Under Minimal Information: Hierarchical Component Selection from One-Bit Feedback
Paper • 2605.00921 • Published
Information Requirements for Service Allocation and Aggregate Verification
Paper • 2604.26808 • Published
Task-Oriented Quantization for Quadratic Scheduling: Centroid Water-Filling and Power-Diagram Encoders
Paper • 2609.20882 • Published
Multi-Resolution Attribution from Adaptive Routing State
Paper • 2605.22866 • Published
Source Identification Is Not Fitness Testing: Measuring the Limits of Synthetic-Data Attribution
Paper • 2610.00417 • Published • 1