Skip to content

Author

Veljko Skarich

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Oct 2026

Poison Valleys: Diagnosing Path-Dependent Failure in LLM Code Repair

Poison Valleys defines plausible but incorrect software mechanisms, place each in a model's hypothesis, implementation plan, or both, and evaluates the generated code with an Oracle and targeted diagnostic coordinates, creating a measured connection from upstream misconception to code mechanism and failure form.

Veljko Skarich · 0 citations
Preprint Aug 2026

Multi-Agent Debate for Explainable Trading: Reasoning, Consensus, and Performance in Simulated Markets

It is suggested that multi-agent debate is most valuable when it preserves independent informational signals rather than simply improving measured reasoning quality, and that interventions enforcing stronger causal reasoning do not improve financial performance.

Ju-Li Huang, A. Alrassan, D. Harischandra et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.