3 papers
cs.LG2026
The Mechanism Matters: When Knowledge Graphs Help Reinforcement Learning
Mohammed Sameer Syed
Knowledge graphs (KGs) are widely used to inject prior knowledge into reinforcement learning (RL), yet the literature is dominated by single-domain, positive-result method papers,…
cs.LG2026
Same Payload, Different Channel: Measuring Trust Asymmetry in Tool-Using Language Models
Mohammed Sameer Syed, Rozhin Yasaei
As language models take on agentic roles that span calling external APIs, reading tool outputs, and acting on instructions embedded in third-party content, their attack surface exp…
cs.LG2026
ShifaMind: A Multiplicative Concept Bottleneck for Interpretable ICD-10 Coding
Mohammed Sameer Syed, Xuan Lu
Automated ICD-10 coding from clinical discharge summaries requires models that are both accurate on long-tailed multi-label classification tasks and interpretable to clinicians. Co…