Conversations Risk Detection LLMs in Financial Agents via Multi-Stage Generative Rollout

• URL: http://arxiv.org/abs/2604.09056v1
• Date: Fri, 10 Apr 2026 07:29:39 GMT
• Title: Conversations Risk Detection LLMs in Financial Agents via Multi-Stage Generative Rollout
• Abstract summary: FinSec is a four-tier security detection framework for financial agent.<n>It enables structured, interpretable, and end-to-end identification of actual financial risks.<n>It significantly enhances the robustness of high-risk dialogue detection.
• Score: 7.95412748755362
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: With the rapid adoption of large language models (LLMs) in financial service scenarios, dialogue security detection under high regulatory risk presents significant challenges. Existing methods mainly rely on single-dimensional semantic judgments or fixed rules, making them inadequate for handling multi-turn semantic evolution and complex regulatory clauses; moreover, they lack models specifically designed for financial security detection. To address these issues, this paper proposes FinSec, a four-tier security detection framework for financial agent. FinSec enables structured, interpretable, and end-to-end identification of actual financial risks, incorporating suspicious behavior pattern analysis, delayed risk and adversarial inference, semantic security analysis, and integrated risk-based decision-making. Notably, FinSec significantly enhances the robustness of high-risk dialogue detection while maintaining model utility. Experimental results demonstrate FinSec's leading performance. In terms of overall detection capability, FinSec achieves an F1 score of 90.13%, improving upon baseline models by 6--14 percentage points; its ASR is reduced to 9.09%, markedly lowering the probability of unsafe outputs; and the AUPRC increases to 0.9189 -- an approximate 9.7% gain over general frameworks. Additionally, in balancing utility and safety, FinSec obtains a composite score of 0.9098, delivering robust and efficient protection for financial agent dialogues.

Related papers

• Herculean: An Agentic Benchmark for Financial Intelligence [114.33879935277217]
  We introduce Herculean, the first skilled benchmark for agentic financial intelligence spanning four representative skill sets.<n>Across frontier agents, we find agents perform relatively well on Trading and Market Insights, but struggle substantially on Hedging and Auditing.<n>Overall, our results point to a key gap in current agents in turning financial reasoning into dependable workflow execution.
  arXiv  Detail & Related papers  (2026-05-14T04:30:49Z)
• Risk-Adjusted Harm Scoring for Automated Red Teaming for LLMs in Financial Services [0.0]
  We propose a risk-aware evaluation framework for security failures in Banking, Financial Services, and Insurance.<n>We combine a domain-specific taxonomy of financial harms, an automated multi-round red-teaming pipeline, and an ensemble-based judging protocol.
  arXiv  Detail & Related papers  (2026-03-11T14:14:13Z)
• Conformal Thinking: Risk Control for Reasoning on a Compute Budget [60.65072883773352]
  Reasoning Large Language Models (LLMs) enable test-time scaling, with dataset-level accuracy improving as the token budget increases.<n>We re-frame the budget setting problem as risk control, limiting the error rate while minimizing compute.<n>Our framework introduces an upper threshold that stops reasoning when the model is confident and a novel lower threshold that preemptively stops unsolvable instances.
  arXiv  Detail & Related papers  (2026-02-03T18:17:22Z)
• FinVault: Benchmarking Financial Agent Safety in Execution-Grounded Environments [29.19862712902908]
  FinVault is the first execution-grounded security benchmark for financial agents.<n>We show that existing defense mechanisms remain ineffective in realistic financial agent settings.
  arXiv  Detail & Related papers  (2026-01-09T03:25:45Z)
• Uni-FinLLM: A Unified Multimodal Large Language Model with Modular Task Heads for Micro-Level Stock Prediction and Macro-Level Systemic Risk Assessment [6.015507338546882]
  Financial institutions and regulators require systems that integrate heterogeneous data to assess risks from stock fluctuations to systemic vulnerabilities.<n>We propose Uni-FinLLM, a unified multimodal large language model that uses a shared Transformer backbone and modular task heads to jointly process financial text, numerical time series, fundamentals, and visual data.
  arXiv  Detail & Related papers  (2026-01-06T03:22:51Z)
• SafeRBench: A Comprehensive Benchmark for Safety Assessment in Large Reasoning Models [60.8821834954637]
  We present SafeRBench, the first benchmark that assesses LRM safety end-to-end.<n>We pioneer the incorporation of risk categories and levels into input design.<n>We introduce a micro-thought chunking mechanism to segment long reasoning traces into semantically coherent units.
  arXiv  Detail & Related papers  (2025-11-19T06:46:33Z)
• Uncovering the Vulnerability of Large Language Models in the Financial Domain via Risk Concealment [29.36824550283463]
  Large Language Models (LLMs) are increasingly integrated into financial applications, yet existing red-teaming research primarily targets harmful content.<n>We introduce Risk-Concealment Attacks (RCA), a novel multi-turn framework that iteratively conceals regulatory risks to provoke seemingly compliant yet regulatory-violating responses.<n>Experiments on FIN-Bench demonstrate that RCA effectively bypasses nine mainstream LLMs, achieving an average attack success rate (ASR) of 93.18%, including 98.28% on GPT-4.1 and 97.56% on OpenAI o1.
  arXiv  Detail & Related papers  (2025-09-07T22:35:15Z)
• Agentar-Fin-R1: Enhancing Financial Intelligence through Domain Expertise, Training Efficiency, and Advanced Reasoning [12.548390779247987]
  We introduce the Agentar-Fin-R1 series of financial large language models.<n>Our optimization approach integrates a high-quality, systematic financial task label system.<n>Our models undergo comprehensive evaluation on mainstream financial benchmarks.
  arXiv  Detail & Related papers  (2025-07-22T17:52:16Z)
• FinHEAR: Human Expertise and Adaptive Risk-Aware Temporal Reasoning for Financial Decision-Making [58.04602111184477]
  FinHEAR is a framework for Human Expertise and Adaptive Risk-aware reasoning.<n>It orchestrates specialized agents to analyze historical trends, interpret current events, and retrieve expert-informed precedents.<n> Empirical results on financial datasets show that FinHEAR consistently outperforms strong baselines across trend prediction and trading tasks.
  arXiv  Detail & Related papers  (2025-06-10T04:06:51Z)
• Agent-SafetyBench: Evaluating the Safety of LLM Agents [72.92604341646691]
  We introduce Agent-SafetyBench, a benchmark designed to evaluate the safety of large language models (LLMs)<n>Agent-SafetyBench encompasses 349 interaction environments and 2,000 test cases, evaluating 8 categories of safety risks and covering 10 common failure modes frequently encountered in unsafe interactions.<n>Our evaluation of 16 popular LLM agents reveals a concerning result: none of the agents achieves a safety score above 60%.
  arXiv  Detail & Related papers  (2024-12-19T02:35:15Z)
• ASSERT: Automated Safety Scenario Red Teaming for Evaluating the Robustness of Large Language Models [65.79770974145983]
  ASSERT, Automated Safety Scenario Red Teaming, consists of three methods -- semantically aligned augmentation, target bootstrapping, and adversarial knowledge injection. We partition our prompts into four safety domains for a fine-grained analysis of how the domain affects model performance. We find statistically significant performance differences of up to 11% in absolute classification accuracy among semantically related scenarios and error rates of up to 19% absolute error in zero-shot adversarial settings.
  arXiv  Detail & Related papers  (2023-10-14T17:10:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.