paperAugust 2026Unreviewed

BiasMix-Finance: Post-Generation KYC Guardrails for LLM Portfolio Advice

Gaurav Kukreja, Parul Kukreja, Mohammed Abraar, R. Dandekar, R. Dandekar, Sreedath Panat

Abstract

Large language models (LLMs) can generate plausible-sounding ETF portfolios while silently violating basic KYC-style constraints on risk, fees, and diversification. This is especially problematic in agentic multi-turn advisory systems, where each draft recommendation can become an action unless guarded by an auditable enforcement layer. We study a model-agnostic, asset-agnostic post-generation guardrail pipeline: (i) enforce a strict JSON allocation schema, (ii) validate allocations against nume

Cite

@misc{kukreja2026biasmixfinance,
  title = {{BiasMix-Finance: Post-Generation KYC Guardrails for LLM Portfolio Advice}},
  author = {Gaurav Kukreja and Parul Kukreja and Mohammed Abraar and R. Dandekar and R. Dandekar and Sreedath Panat},
  year = {2026},
  month = aug,
  eprint = {2608.28646},
  archivePrefix = {arXiv},
  url = {https://www.semanticscholar.org/paper/35614c01d076e0ee928e20ec6fa9652abed6457b}
}