Related papers: Disentangling Deception and Hallucination Failures in LLMs

Related papers

Dissociating the Internal Representations of Sycophancy in LLMs [9.383958408772694]
Large Language Models (LLMs) frequently exhibit sycophancy, where they agree with a user's statement even when incorrect.<n>We train linear probes and construct steering vectors on activations of one subtype and evaluate their transfer to the other subtype to measure to what extent they share representations.<n>We find evidence that different LLMs represent these subtypes differently, with either more unified or more distinct and causally interfering representations.
arXiv Detail & Related papers (2026-07-08T04:54:54Z)
Why LLMs Hallucinate on Structured Knowledge: A Mechanistic Analysis of Reasoning over Linearized Representations [36.45654492179688]
We find that hallucinations arise from systematic internal dynamics rather than random noise.<n>First, attention disproportionately concentrates toward shortcut-like structural cues rather than distributing across the full context.<n>Second, feed-forward representations fail to ground the provided knowledge, causing the model to revert to parametric memory.
arXiv Detail & Related papers (2026-05-25T22:08:59Z)
Decomposing and Steering Functional Metacognition in Large Language Models [7.62618887128252]
We show that large language models maintain a decomposable space of functional metacognitive states.<n>We show that these states are linearly decodable from internal activations and exhibit distinct layer-wise profiles.<n>Our findings suggest that benchmark performance reflects not only task competence but also the activation of specific functional metacognitive states.
arXiv Detail & Related papers (2026-05-09T13:22:47Z)
From Architecture to Output: Structural Origins of Hallucination in Large Language Models and the Amplifying Role of Data [0.0]
This paper analyses hallucination as a structural consequence of three architectural decisions that together form a compound failure system.<n>Self-attention's co-occurrence learning substitutes statistical proximity for semantic meaning and produces entity confusion.<n>Autoregressive decoding's permanent left-to-right commitment under exposure bias ensures that a single wrong token cascades forward through the entire output sequence without revision.
arXiv Detail & Related papers (2026-04-29T11:34:09Z)
Anatomy of a Lie: A Multi-Stage Diagnostic Framework for Tracing Hallucinations in Vision-Language Models [62.932580559941414]
Vision-Language Models (VLMs) frequently "hallucinate" - generate plausible yet factually incorrect statements.<n>We propose a new paradigm for diagnosing hallucinations, recasting them from static output errors into dynamic pathologies of a model's computational cognition.
arXiv Detail & Related papers (2026-03-16T17:20:38Z)
Causality is Key for Interpretability Claims to Generalise [35.833847356014154]
Interpretability research on large language models (LLMs) has yielded important insights into model behaviour.<n> recurring pitfalls persist: findings that do not generalise, and causal interpretations that outrun the evidence.<n>Pearl's causal hierarchy clarifies what an interpretability study can justify.
arXiv Detail & Related papers (2026-02-18T18:45:04Z)
Operational Noncommutativity in Sequential Metacognitive Judgments [0.0]
We show that metacognitive evaluations are modeled as state-transforming operations acting on an internal state space with probabilistic readouts.<n>We also outline a behavioral paradigm involving sequential confidence, error-likelihood, and feeling-of-knowing judgments following a perceptual decision.
arXiv Detail & Related papers (2026-02-15T08:15:48Z)
Detecting Misbehaviors of Large Vision-Language Models by Evidential Uncertainty Quantification [27.02252748004729]
Large vision-language models (LVLMs) have shown substantial advances in multimodal understanding and generation.<n>They frequently produce unreliable or even harmful content, such as fact hallucinations or dangerous instructions.<n>Evidential Uncertainty Quantification (EUQ) captures both information conflict and ignorance for effective detection of LVLM misbehaviors.
arXiv Detail & Related papers (2026-02-05T10:51:39Z)
Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts [74.47786985522762]
We identify a critical failure mode termed textual inertia, where models tend to blindly adhere to the erroneous text while neglecting conflicting visual evidence.<n>We propose the LogicGraph Perturbation Protocol that structurally injects perturbations into the reasoning chains of diverse LMMs.<n>Results reveal that models successfully self-correct in less than 10% of cases and predominantly succumb to blind textual error propagation.
arXiv Detail & Related papers (2026-01-07T16:39:34Z)
Fantastic Reasoning Behaviors and Where to Find Them: Unsupervised Discovery of the Reasoning Process [66.38541693477181]
We propose an unsupervised framework for discovering reasoning vectors, which we define as directions in the activation space that encode distinct reasoning behaviors.<n>By segmenting chain-of-thought traces into sentence-level'steps', we uncover disentangled features corresponding to interpretable behaviors such as reflection and backtracking.<n>We demonstrate the ability to control response confidence by identifying confidence-related vectors in the SAE decoder space.
arXiv Detail & Related papers (2025-12-30T05:09:11Z)
Cognitive Foundations for Reasoning and Their Manifestation in LLMs [63.12951576410617]
Large language models (LLMs) solve complex problems yet fail on simpler variants, suggesting they achieve correct outputs through mechanisms fundamentally different from human reasoning.<n>We synthesize cognitive science research into a taxonomy of 28 cognitive elements spanning reasoning invariants, meta-cognitive controls, representations for organizing reasoning & knowledge, and transformation operations.<n>We develop test-time reasoning guidance that automatically scaffold successful structures, improving performance by up to 66.7% on complex problems.
arXiv Detail & Related papers (2025-11-20T18:59:00Z)
Self-Correcting Large Language Models: Generation vs. Multiple Choice [29.697851249014192]
Large language models have recently demonstrated remarkable abilities to self-correct their responses through iterative refinement.<n>We compare performance trends and error-correction behaviors across various natural language understanding and reasoning tasks.<n>Our findings highlight that the design of self-correction mechanisms should take into account the interaction between task structure and output space.
arXiv Detail & Related papers (2025-11-12T14:46:40Z)
Distributional Semantics Tracing: A Framework for Explaining Hallucinations in Large Language Models [4.946483489399819]
Large Language Models (LLMs) are prone to hallucination, the generation of factually incorrect statements.<n>This work investigates the intrinsic, architectural origins of this failure mode through three primary contributions.
arXiv Detail & Related papers (2025-10-07T16:40:31Z)
Investigating VLM Hallucination from a Cognitive Psychology Perspective: A First Step Toward Interpretation with Intriguing Observations [60.63340688538124]
Hallucination is a long-standing problem that has been actively investigated in Vision-Language Models (VLMs)<n>Existing research commonly attributes hallucinations to technical limitations or sycophancy bias, where the latter means the models tend to generate incorrect answers to align with user expectations.<n>In this work, we introduce a psychological taxonomy, categorizing VLMs' cognitive biases that lead to hallucinations, including sycophancy, logical inconsistency, and a newly identified VLMs behaviour: appeal to authority.
arXiv Detail & Related papers (2025-07-03T19:03:16Z)
Causality can systematically address the monsters under the bench(marks) [64.36592889550431]
Benchmarks are plagued by various biases, artifacts, or leakage.<n>Models may behave unreliably due to poorly explored failure modes.<n> causality offers an ideal framework to systematically address these challenges.
arXiv Detail & Related papers (2025-02-07T17:01:37Z)
Failure Modes of LLMs for Causal Reasoning on Narratives [51.19592551510628]
We investigate the interaction between world knowledge and logical reasoning.<n>We find that state-of-the-art large language models (LLMs) often rely on superficial generalizations.<n>We show that simple reformulations of the task can elicit more robust reasoning behavior.
arXiv Detail & Related papers (2024-10-31T12:48:58Z)
Self-correction is Not An Innate Capability in Large Language Models [13.268938380591765]
We investigate the underlying mechanism of moral self-correction by addressing a fundamental question: is moral self-correction an innate capability of LLMs?<n>We show that moral self-correction is not an inherent capability of LLMs, as they are neither morally sensitive nor able to effectively incorporate external feedback during the self-correction process.
arXiv Detail & Related papers (2024-10-27T16:52:21Z)
Feedback in Imitation Learning: Confusion on Causality and Covariate Shift [12.93527098342393]
We argue that conditioning policies on previous actions leads to a dramatic divergence between "held out" error and performance of the learner in situ. We analyze existing benchmarks used to test imitation learning approaches. We find, in a surprising contrast with previous literature, that naive behavioral cloning provides excellent results.
arXiv Detail & Related papers (2021-02-04T20:18:56Z)

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