One Size Fits None: Heuristic Collapse in LLM Investment Advice

• URL: http://arxiv.org/abs/2604.23837v1
• Date: Sun, 26 Apr 2026 18:45:11 GMT
• Title: One Size Fits None: Heuristic Collapse in LLM Investment Advice
• Abstract summary: Large language models are increasingly deployed as advisors in high-stakes domains.<n>We investigate whether frontier LLMs actually do this, or whether they instead exhibit collapse.<n>We find that investment allocation decisions are largely determined by self-reported risk tolerance, while other relevant factors contribute minimally.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models are increasingly deployed as advisors in high-stakes domains -- answering medical questions, interpreting legal documents, recommending financial products -- where good advice requires integrating a user's full context rather than responding to salient surface features. We investigate whether frontier LLMs actually do this, or whether they instead exhibit heuristic collapse: a systematic reduction of complex, multi-factor decisions to a small number of dominant inputs. We study the phenomenon in investment advice, where legal standards explicitly require individualized reasoning over a client's full circumstances. Applying interpretable surrogate models to LLM outputs, we find systematic heuristic collapse: investment allocation decisions are largely determined by self-reported risk tolerance, while other relevant factors contribute minimally. We further find that web search partially attenuates heuristic collapse but does not resolve it. These findings suggest that heuristic collapse is not resolved by web search augmentation or model scale alone, and that deploying LLMs as advisors requires auditing input sensitivity, not just output quality.

Related papers

• Your AI, On a Dial: Controlling Investment Bias in LLMs with a Single Neuron [76.4407829130164]
  We study whether a model's overall investment stance can be calibrated to a specified direction and strength.<n>We introduce an investment-bias dial, an inference-time intervention on a single neuron.<n>We find that the dial produces monotonic changes in investment stance without modifying prompts or model parameters.
  arXiv  Detail & Related papers  (2026-08-24T06:33:02Z)
• A2RBench: An Automatic Paradigm for Formally Verifiable Abstract Reasoning Benchmark Generation [59.98516959731531]
  Abstract reasoning ability reflects intelligence and capacity of LLMs to extract and apply abstract rules.<n>Existing benchmarks either rely on expensive manual annotation, limiting their scale, or risk measuring memorization rather than genuine reasoning.<n>We introduce an automated pipeline named A2RBench, encompassing generation, expansion, evaluation, and analysis.
  arXiv  Detail & Related papers  (2026-05-17T06:14:20Z)
• When Do We Need LLMs? A Diagnostic for Language-Driven Bandits [44.62361611699998]
  We study Contextual Multi-Armed Bandits (CMABs) for non-episodic sequential decision making problems in finance.<n>Our results provide a principled deployment framework for cost-effective, uncertainty-aware decision systems.
  arXiv  Detail & Related papers  (2026-04-07T13:20:06Z)
• Reasoning with Confidence: Efficient Verification of LLM Reasoning Steps via Uncertainty Heads [104.9566359759396]
  We propose a lightweight alternative for step-level reasoning verification based on data-driven uncertainty scores.<n>Our findings suggest that the internal states of LLMs encode their uncertainty and can serve as reliable signals for reasoning verification.
  arXiv  Detail & Related papers  (2025-11-09T03:38:29Z)
• FinLFQA: Evaluating Attributed Text Generation of LLMs in Financial Long-Form Question Answering [57.43420753842626]
  FinLFQA is a benchmark designed to evaluate the ability of Large Language Models to generate long-form answers to complex financial questions.<n>We provide an automatic evaluation framework covering both answer quality and attribution quality.
  arXiv  Detail & Related papers  (2025-10-07T20:06:15Z)
• LLM Enhancement with Domain Expert Mental Model to Reduce LLM Hallucination with Causal Prompt Engineering [0.3437656066916039]
  We propose a technology based on optimized human-machine dialogue and monotone Boolean and k-valued functions to discover a computationally tractable personal expert mental model (EMM) of decision-making.<n>Our EMM algorithm for LLM prompt engineering has four steps: factor identification, (2) hierarchical structuring of factors, (3) generating a generalized expert mental model specification, and (4) generating a detailed generalized expert mental model from that specification.
  arXiv  Detail & Related papers  (2025-09-13T14:35:51Z)
• Your AI, Not Your View: The Bias of LLMs in Investment Analysis [62.388554963415906]
  In finance, Large Language Models (LLMs) face frequent knowledge conflicts arising from discrepancies between their pre-trained parametric knowledge and real-time market data.<n>These conflicts are especially problematic in real-world investment services, where a model's inherent biases can misalign with institutional objectives.<n>We propose an experimental framework to investigate emergent behaviors in such conflict scenarios, offering a quantitative analysis of bias in investment analysis.
  arXiv  Detail & Related papers  (2025-07-28T16:09:38Z)
• Answer, Refuse, or Guess? Investigating Risk-Aware Decision Making in Language Models [63.559461750135334]
  Language models (LMs) are increasingly used to build agents that can act autonomously to achieve goals.<n>We study this "answer-or-defer" problem with an evaluation framework that systematically varies human-specified risk structures.<n>We find that a simple skill-decomposition method, which isolates the independent skills required for answer-or-defer decision making, can consistently improve LMs' decision policies.
  arXiv  Detail & Related papers  (2025-03-03T09:16:26Z)
• mR$^2$AG: Multimodal Retrieval-Reflection-Augmented Generation for Knowledge-Based VQA [78.45521005703958]
  multimodal Retrieval-Augmented Generation (mRAG) is naturally introduced to provide MLLMs with comprehensive and up-to-date knowledge. We propose a novel framework called textbfRetrieval-textbfReftextbfAugmented textbfGeneration (mR$2$AG) which achieves adaptive retrieval and useful information localization. mR$2$AG significantly outperforms state-of-the-art MLLMs on INFOSEEK and Encyclopedic-VQA
  arXiv  Detail & Related papers  (2024-11-22T16:15:50Z)
• DeFine: Decision-Making with Analogical Reasoning over Factor Profiles [35.9909472797192]
  textscDeFine is a modular framework that constructs probabilistic factor profiles from complex scenarios.<n>It then integrates these profiles with analogical reasoning to guide LLMs in making critical decisions in new situations.<n>This approach is particularly useful in areas such as consulting and financial deliberation, where making decisions under uncertainty is vital.
  arXiv  Detail & Related papers  (2024-10-02T17:29:34Z)
• Assessing Hidden Risks of LLMs: An Empirical Study on Robustness, Consistency, and Credibility [37.682136465784254]
  We conduct over a million queries to the mainstream large language models (LLMs) including ChatGPT, LLaMA, and OPT. We find that ChatGPT is still capable to yield the correct answer even when the input is polluted at an extreme level. We propose a novel index associated with a dataset that roughly decides the feasibility of using such data for LLM-involved evaluation.
  arXiv  Detail & Related papers  (2023-05-15T15:44:51Z)

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.