Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs

• URL: http://arxiv.org/abs/2506.21561v1
• Date: Thu, 12 Jun 2025 00:19:36 GMT
• Title: Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs
• Authors: Emilio Barkett, Olivia Long, Madhavendra Thakur,
• Abstract summary: This study presents the largest evaluation to date of large language models' veracity detection capabilities.<n>Rates of truth-bias, or the likelihood to believe a statement is true, are lower in reasoning models than in non-reasoning models.<n>Most concerning, we identify sycophantic tendencies in several advanced models.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Despite their widespread use in fact-checking, moderation, and high-stakes decision-making, large language models (LLMs) remain poorly understood as judges of truth. This study presents the largest evaluation to date of LLMs' veracity detection capabilities and the first analysis of these capabilities in reasoning models. We had eight LLMs make 4,800 veracity judgments across several prompts, comparing reasoning and non-reasoning models. We find that rates of truth-bias, or the likelihood to believe a statement is true, regardless of whether it is actually true, are lower in reasoning models than in non-reasoning models, but still higher than human benchmarks. Most concerning, we identify sycophantic tendencies in several advanced models (o4-mini and GPT-4.1 from OpenAI, R1 from DeepSeek), which displayed an asymmetry in detection accuracy, performing well in truth accuracy but poorly in deception accuracy. This suggests that capability advances alone do not resolve fundamental veracity detection challenges in LLMs.

Related papers

• WakenLLM: Evaluating Reasoning Potential and Stability in LLMs via Fine-Grained Benchmarking [14.76224690767612]
  Large Language Models (LLMs) frequently output the label Unknown in reasoning tasks.<n>We introduce WakenLLM, a framework that quantifies the portion of Unknown output attributable to model incapacity.
  arXiv  Detail & Related papers  (2025-07-22T03:21:48Z)
• The Trilemma of Truth in Large Language Models [1.62933895796838]
  We examine two common methods for probing the veracity of large language models (LLMs)<n>We introduce sAwMIL, a probing method that utilizes the internal activations of LLMs to separate statements into true, false, and neither.<n>We evaluate sAwMIL on 5 validity criteria across 16 open-source LLMs, including both default and chat-based variants, as well as on 3 new datasets.
  arXiv  Detail & Related papers  (2025-06-30T14:49:28Z)
• Verifying the Verifiers: Unveiling Pitfalls and Potentials in Fact Verifiers [59.168391398830515]
  We evaluate 12 pre-trained LLMs and one specialized fact-verifier, using a collection of examples from 14 fact-checking benchmarks.<n>We highlight the importance of addressing annotation errors and ambiguity in datasets.<n> frontier LLMs with few-shot in-context examples, often overlooked in previous works, achieve top-tier performance.
  arXiv  Detail & Related papers  (2025-06-16T10:32:10Z)
• Don't Take the Premise for Granted: Evaluating the Premise Critique Ability of Large Language Models [11.379764847748378]
  Large language models (LLMs) often uncritically accept flawed or contradictory premises, leading to inefficient reasoning and unreliable outputs.<n>This emphasizes the significance of possessing the textbfPremise Critique Ability for LLMs, defined as the capacity to proactively identify and articulate errors in input premises.<n>We introduce the textbfPremise Critique Bench (PCBench), designed by incorporating four error types across three difficulty levels, paired with multi-faceted evaluation metrics.
  arXiv  Detail & Related papers  (2025-05-29T17:49:44Z)
• Have Large Language Models Learned to Reason? A Characterization via 3-SAT Phase Transition [11.422434149376478]
  Large Language Models (LLMs) have been touted as AI models possessing advanced reasoning abilities.<n>In theory, autoregressive LLMs with Chain-of-Thought (CoT) can perform more serial computations to solve complex reasoning tasks.<n>Recent studies suggest that, despite this capacity, LLMs do not truly learn to reason but instead fit on statistical features.
  arXiv  Detail & Related papers  (2025-04-04T20:57:36Z)
• The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems [22.458311369795112]
  We introduce a large-scale human-collected dataset for measuring honesty directly.<n>We find that while larger models obtain higher accuracy on our benchmark, they do not become more honest.<n>We find that simple methods, such as representation engineering interventions, can improve honesty.
  arXiv  Detail & Related papers  (2025-03-05T18:59:23Z)
• Understanding Knowledge Drift in LLMs through Misinformation [11.605377799885238]
  Large Language Models (LLMs) have revolutionized numerous applications, making them an integral part of our digital ecosystem. We analyze the susceptibility of state-of-the-art LLMs to factual inaccuracies when they encounter false information in a QnA scenario. Our experiments reveal that an LLM's uncertainty can increase up to 56.6% when the question is answered incorrectly.
  arXiv  Detail & Related papers  (2024-09-11T08:11:16Z)
• A Peek into Token Bias: Large Language Models Are Not Yet Genuine Reasoners [58.15511660018742]
  This study introduces a hypothesis-testing framework to assess whether large language models (LLMs) possess genuine reasoning abilities. We develop carefully controlled synthetic datasets, featuring conjunction fallacy and syllogistic problems.
  arXiv  Detail & Related papers  (2024-06-16T19:22:53Z)
• Large Language Models Must Be Taught to Know What They Don't Know [97.90008709512921]
  We show that fine-tuning on a small dataset of correct and incorrect answers can create an uncertainty estimate with good generalization and small computational overhead.<n>We also investigate the mechanisms that enable reliable uncertainty estimation, finding that many models can be used as general-purpose uncertainty estimators.
  arXiv  Detail & Related papers  (2024-06-12T16:41:31Z)
• How Easy is It to Fool Your Multimodal LLMs? An Empirical Analysis on Deceptive Prompts [54.07541591018305]
  We present MAD-Bench, a benchmark that contains 1000 test samples divided into 5 categories, such as non-existent objects, count of objects, and spatial relationship. We provide a comprehensive analysis of popular MLLMs, ranging from GPT-4v, Reka, Gemini-Pro, to open-sourced models, such as LLaVA-NeXT and MiniCPM-Llama3. While GPT-4o achieves 82.82% accuracy on MAD-Bench, the accuracy of any other model in our experiments ranges from 9% to 50%.
  arXiv  Detail & Related papers  (2024-02-20T18:31:27Z)
• 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)
• A Closer Look at the Self-Verification Abilities of Large Language Models in Logical Reasoning [73.77088902676306]
  We take a closer look at the self-verification abilities of large language models (LLMs) in the context of logical reasoning. Our main findings suggest that existing LLMs could struggle to identify fallacious reasoning steps accurately and may fall short of guaranteeing the validity of self-verification methods.
  arXiv  Detail & Related papers  (2023-11-14T07:13:10Z)

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.