Reasoning and the Trusting Behavior of DeepSeek and GPT: An Experiment Revealing Hidden Fault Lines in Large Language Models

• URL: http://arxiv.org/abs/2502.12825v2
• Date: Wed, 19 Feb 2025 11:57:19 GMT
• Title: Reasoning and the Trusting Behavior of DeepSeek and GPT: An Experiment Revealing Hidden Fault Lines in Large Language Models
• Authors: Rubing Li, João Sedoc, Arun Sundararajan,
• Abstract summary: Low perceived switching frictions can lead to choices that do not consider more subtle behavior changes.<n>Our experiments use a popular game-theoretic behavioral economics model of trust to show stark differences in the trusting behavior of OpenAI's and DeepSeek's models.
• Score: 7.463303856292452
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: When encountering increasingly frequent performance improvements or cost reductions from a new large language model (LLM), developers of applications leveraging LLMs must decide whether to take advantage of these improvements or stay with older tried-and-tested models. Low perceived switching frictions can lead to choices that do not consider more subtle behavior changes that the transition may induce. Our experiments use a popular game-theoretic behavioral economics model of trust to show stark differences in the trusting behavior of OpenAI's and DeepSeek's models. We highlight a collapse in the economic trust behavior of the o1-mini and o3-mini models as they reconcile profit-maximizing and risk-seeking with future returns from trust, and contrast it with DeepSeek's more sophisticated and profitable trusting behavior that stems from an ability to incorporate deeper concepts like forward planning and theory-of-mind. As LLMs form the basis for high-stakes commercial systems, our results highlight the perils of relying on LLM performance benchmarks that are too narrowly defined and suggest that careful analysis of their hidden fault lines should be part of any organization's AI strategy.

Related papers

• Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling [49.41422138354821]
  We propose a principled reward modeling framework that integrates non-negative factor analysis into the Bradley-Terry preference model.<n>BNRM represents rewards through a sparse, non-negative latent factor generative process.<n>We show that BNRM substantially mitigates reward over-optimization, improves robustness under distribution shifts, and yields more interpretable reward decompositions than strong baselines.
  arXiv  Detail & Related papers  (2026-02-11T08:14:11Z)
• Evaluating GRPO and DPO for Faithful Chain-of-Thought Reasoning in LLMs [0.7045900712659982]
  Chain-of-thought (CoT) reasoning has emerged as a powerful technique for improving the problem-solving capabilities of large language models (LLMs)
  arXiv  Detail & Related papers  (2025-12-27T16:07:00Z)
• Large Language Newsvendor: Decision Biases and Cognitive Mechanisms [2.7070404673380817]
  Large language models (LLMs) are increasingly integrated into business decision making.<n>LLMs replicate and amplify human cognitive biases.<n>This is particularly critical in high-stakes operational contexts like supply chain management.
  arXiv  Detail & Related papers  (2025-12-14T04:51:53Z)
• Interpreting LLMs as Credit Risk Classifiers: Do Their Feature Explanations Align with Classical ML? [4.0057196015831495]
  Large Language Models (LLMs) are increasingly explored as flexible alternatives to classical machine learning models for classification tasks through zero-shot prompting.<n>This study conducts a systematic comparison between zero-shot LLM-based classifiers and LightGBM, a state-of-the-art gradient-boosting model, on a real-world loan default prediction task.<n>We evaluate their predictive performance, analyze feature attributions using SHAP, and assess the reliability of LLM-generated self-explanations.
  arXiv  Detail & Related papers  (2025-10-29T17:05:00Z)
• Shop-R1: Rewarding LLMs to Simulate Human Behavior in Online Shopping via Reinforcement Learning [27.226155951073064]
  Shop-R1 is a novel reinforcement learning framework aimed at enhancing the reasoning ability of Large Language Models (LLMs)<n>For rationale generation, we leverage internal model signals (e.g., logit distributions) to guide the reasoning process in a self-supervised manner.<n>For action prediction, we propose a hierarchical reward structure with difficulty-aware scaling to prevent reward hacking.
  arXiv  Detail & Related papers  (2025-07-23T18:10:43Z)
• One Token to Fool LLM-as-a-Judge [52.45386385722788]
  Large language models (LLMs) are increasingly trusted as automated judges, assisting evaluation and providing reward signals for training other models.<n>We uncover a critical vulnerability even in this reference-based paradigm: generative reward models are systematically susceptible to reward hacking.
  arXiv  Detail & Related papers  (2025-07-11T17:55:22Z)
• Mitigating Hidden Confounding by Progressive Confounder Imputation via Large Language Models [46.92706900119399]
  We make the first attempt to mitigate hidden confounding using large language models (LLMs)<n>We propose ProCI, a framework that elicits the semantic and world knowledge of LLMs to iteratively generate, impute, and validate hidden confounders.<n>Extensive experiments demonstrate that ProCI uncovers meaningful confounders and significantly improves treatment effect estimation.
  arXiv  Detail & Related papers  (2025-06-26T03:49:13Z)
• A biologically Inspired Trust Model for Open Multi-Agent Systems that is Resilient to Rapid Performance Fluctuations [0.0]
  Existing trust models face challenges related to agent mobility, changing behaviors, and the cold start problem. We introduce a biologically inspired trust model in which trustees assess their own capabilities and store trust data locally. This design improves mobility support, reduces communication overhead, resists disinformation, and preserves privacy.
  arXiv  Detail & Related papers  (2025-04-17T08:21:54Z)
• Supervised Optimism Correction: Be Confident When LLMs Are Sure [91.7459076316849]
  We establish a novel theoretical connection between supervised fine-tuning and offline reinforcement learning. We show that the widely used beam search method suffers from unacceptable over-optimism. We propose Supervised Optimism Correction, which introduces a simple yet effective auxiliary loss for token-level $Q$-value estimations.
  arXiv  Detail & Related papers  (2025-04-10T07:50:03Z)
• Adversarial Reasoning at Jailbreaking Time [49.70772424278124]
  We develop an adversarial reasoning approach to automatic jailbreaking via test-time computation.<n>Our approach introduces a new paradigm in understanding LLM vulnerabilities, laying the foundation for the development of more robust and trustworthy AI systems.
  arXiv  Detail & Related papers  (2025-02-03T18:59:01Z)
• The Dual-use Dilemma in LLMs: Do Empowering Ethical Capacities Make a Degraded Utility? [54.18519360412294]
  Large Language Models (LLMs) must balance between rejecting harmful requests for safety and accommodating legitimate ones for utility. This paper presents a Direct Preference Optimization (DPO) based alignment framework that achieves better overall performance. We analyze experimental results obtained from testing DeepSeek-R1 on our benchmark and reveal the critical ethical concerns raised by this highly acclaimed model.
  arXiv  Detail & Related papers  (2025-01-20T06:35:01Z)
• Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment [30.605500809158986]
  We propose a novel causal reward modeling approach that integrates causal inference to mitigate spurious correlations.<n>Our approach mitigates various types of spurious correlations effectively, resulting in more reliable and fair alignment of LLMs with human preferences.
  arXiv  Detail & Related papers  (2025-01-16T16:00:37Z)
• OCEAN: Offline Chain-of-thought Evaluation and Alignment in Large Language Models [68.17018458283651]
  This work focuses on the offline evaluation of the chain-of-thought capabilities of LLMs. We use knowledge graphs (e.g., Wikidata5m) to provide feedback on the generated chain of thoughts. We show how to optimize LLMs based on the proposed evaluation method.
  arXiv  Detail & Related papers  (2024-10-31T07:48:44Z)
• Insights from the Inverse: Reconstructing LLM Training Goals Through Inverse RL [7.988692259455583]
  Large language models (LLMs) trained with Reinforcement Learning from Human Feedback have demonstrated remarkable capabilities, but their underlying reward functions and decision-making processes remain opaque. This paper introduces a novel approach to interpreting LLMs by applying inverse reinforcement learning (IRL) to recover their implicit reward functions. We conduct experiments on toxicity-aligned LLMs of varying sizes, extracting reward models that achieve up to 80.40% accuracy in predicting human preferences.
  arXiv  Detail & Related papers  (2024-10-16T12:14:25Z)
• Behavioral Bias of Vision-Language Models: A Behavioral Finance View [1.7205106391379021]
  We study the potential behavioral biases of Large Vision-Language Models (LVLMs) from a behavioral finance perspective. Our evaluations find that open-source LVLMs such as LLaVA-NeXT, MobileVLM-V2, Mini-Gemini, Mini-Llama3-V 2.5 and Phi-3-vision-128k suffer significantly from these two biases. Our observations highlight directions in which open-source models can improve.
  arXiv  Detail & Related papers  (2024-09-23T17:54:47Z)
• Test-Time Fairness and Robustness in Large Language Models [17.758735680493917]
  Frontier Large Language Models (LLMs) can be socially discriminatory or sensitive to spurious features of their inputs. Existing solutions, which instruct the LLM to be fair or robust, rely on the model's implicit understanding of bias. We show that our prompting strategy, unlike implicit instructions, consistently reduces the bias of frontier LLMs.
  arXiv  Detail & Related papers  (2024-06-11T20:05:15Z)
• Are Large Language Models Good Prompt Optimizers? [65.48910201816223]
  We conduct a study to uncover the actual mechanism of LLM-based Prompt Optimization. Our findings reveal that the LLMs struggle to identify the true causes of errors during reflection, tending to be biased by their own prior knowledge. We introduce a new "Automatic Behavior Optimization" paradigm, which directly optimize the target model's behavior in a more controllable manner.
  arXiv  Detail & Related papers  (2024-02-03T09:48:54Z)
• The ART of LLM Refinement: Ask, Refine, and Trust [85.75059530612882]
  We propose a reasoning with refinement objective called ART: Ask, Refine, and Trust. It asks necessary questions to decide when an LLM should refine its output. It achieves a performance gain of +5 points over self-refinement baselines.
  arXiv  Detail & Related papers  (2023-11-14T07:26:32Z)
• Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models [51.3422222472898]
  We document the capability of large language models (LLMs) like ChatGPT to predict stock price movements using news headlines. We develop a theoretical model incorporating information capacity constraints, underreaction, limits-to-arbitrage, and LLMs.
  arXiv  Detail & Related papers  (2023-04-15T19:22:37Z)
• Reparameterized Variational Divergence Minimization for Stable Imitation [57.06909373038396]
  We study the extent to which variations in the choice of probabilistic divergence may yield more performant ILO algorithms. We contribute a re parameterization trick for adversarial imitation learning to alleviate the challenges of the promising $f$-divergence minimization framework. Empirically, we demonstrate that our design choices allow for ILO algorithms that outperform baseline approaches and more closely match expert performance in low-dimensional continuous-control tasks.
  arXiv  Detail & Related papers  (2020-06-18T19:04:09Z)

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.