ARslan Ahmad

ARslan-Ahamd

arslan-ahm
arslan-ahmad-dev

AI & ML interests

None yet

Recent Activity

published a Space 5 days ago

ARslan-Ahamd/graphtrace-bench

published a Space 5 days ago

ARslan-Ahamd/provenance-segment

updated a Space 6 days ago

ARslan-Ahamd/graphtrace-bench

View all activity

Organizations

None yet

spaces 10

GraphTrace-Bench

🚀

Verifiable multi-hop QA over a provenance-linked paper KG

Provenance-Segment

🔎

Check long-document QA claims against their exact sources

Grounded Document Extraction

📄

Extraction where hallucination is unrepresentable

Self-Supervised Log Anomaly Triage

🛡

When a better model of normal is a worse detector

The Latency Budget Nobody Measured

🎙

Labelling who is speaking, live rather than afterwards

Calibrated Uncertainty over Thresholds

🔬

Two pixels at P=0.5 that mean opposite things

View 10 Spaces

models 0

None public yet

datasets 1

ARslan-Ahamd/roman-urdu-retrieval-benchmark

Viewer • Updated Jul 2 • 702 • 16