Hallucination as an Anomaly: Dynamic Intervention via Probabilistic Circuits

• URL: http://arxiv.org/abs/2605.05953v1
• Date: Thu, 07 May 2026 10:02:27 GMT
• Title: Hallucination as an Anomaly: Dynamic Intervention via Probabilistic Circuits
• Abstract summary: We propose PCNET, a Probabilistic Circuit trained as a tractable density estimator over the LLM residual stream.<n>The method detects hallucinations as geometric anomalies on the factual manifold, which is done via exact Negative Log-Likelihood computation.<n>PCNET achieves near-perfect hallucination detection across CoQA, SQuAD v2.0, and TriviaQA, with AUROC reaching up to 99%.
• Score: 6.667596224057802
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: One of the most critical challenges in Large Language Models is their tendency to hallucinate, i.e., produce factually incorrect responses. Existing approaches show promising results in terms of hallucination correction, but still suffer from a main limitation: they apply corrections indiscriminately to every token, corrupting also the originally correct generations. To overcome this drawback, we propose PCNET, a Probabilistic Circuit trained as a tractable density estimator over the LLM residual stream. The method detects hallucinations as geometric anomalies on the factual manifold, which is done via exact Negative Log-Likelihood computation, hence without the need for sampling, external verifiers, or weight modifications, as in existing techniques. To demonstrate its effectiveness, we exploit PCNET as a dynamic gate that distinguishes hallucinated from factual hidden states at each decoding step. This triggers our second main contribution, PC-LDCD (Probabilistic Circuit Latent Density Contrastive Decoding), only when the latent geometry deviates from factual regions, while leaving correct generations untouched. Across four LLMs, ranging from 1B to 8B models, and four benchmarks covering conversational reasoning, knowledge-intensive QA, reading comprehension, and truthfulness, PCNET achieves near-perfect hallucination detection across CoQA, SQuAD v2.0, and TriviaQA, with AUROC reaching up to 99%. Moreover, PC-LDCD obtains the highest True+Info, MC2, and MC3 scores on TruthfulQA in three out of four models, in comparison with state-of-the-art baselines, while reducing the mean corruption rate to 53.7% and achieving a preservation rate of 79.3%. Our proposed method is publicly available on GitHub.

Related papers

• Look Before You Leap: Factual Decoding with Internal Attribution Signals [14.706754790960902]
  Hallucination remains a critical challenge in large language models.<n>We propose DescaPE, a decoding framework that leverages internal model signals to suppress hallucination-prone trajectories at time.
  arXiv  Detail & Related papers  (2026-09-14T15:35:42Z)
• NovGauge: A Fine-Grained Benchmark for Diagnosing LLMs' Capability in Paper Novelty Assessment [54.4265627247104]
  Large language models (LLMs) are increasingly used in peer review at major AI conferences.<n>Existing benchmarks assess novelty as a single holistic score.<n>We present NovGauge, a human-anchored benchmark for fine-grained novelty assessment diagnosis.
  arXiv  Detail & Related papers  (2026-09-10T08:34:56Z)
• Architecture-Dependent Causal Transfer of Activation States Across Large Language Models [51.56484100374058]
  Internal language activation states can be transferred causally between different large language models.<n>Injecting projected activations into a target model during generation produces a statistically significant, preregistered causal effect.<n>We conclude that end-to-end activation-state transfer between LLMs, as currently implemented, is architecture-dependent rather than universal.
  arXiv  Detail & Related papers  (2026-08-17T09:53:27Z)
• Now You (Still) See Me: Detecting Evasive Steganographic Payloads in LLMs [52.149036302760386]
  Large language models can be fine-tuned to encode prompt-borne secrets into fluent, seemingly benign outputs.<n>Recent work proposes detection using linear probes that recover the secret from internal activations.<n>We show that this defense can be systematically evaded, but that detectability can be recovered through a targeted data-level intervention.
  arXiv  Detail & Related papers  (2026-06-08T12:27:11Z)
• Distributional Energy-Based Models for Uncertainty-Aware Structured LLM Reasoning [40.342912574072024]
  Large Language Models produce structured outputs such as travel plans or code solutions.<n>Individual reasoning steps may appear correct while the output as a whole violates budgets, fails test cases, or contradicts earlier deductions.<n>We propose a learned quality scorer with deterministic analytical constraint penalties for verifying structured LLM outputs.
  arXiv  Detail & Related papers  (2026-05-15T17:08:27Z)
• HalluScan: A Systematic Benchmark for Detecting and Mitigating Hallucinations in Instruction-Following LLMs [0.0]
  Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language processing tasks.<n>Yet they remain susceptible to hallucinations -- generating content that is factually incorrect, unfaithful to provided context, or misaligned with user instructions.<n>We present HalluScan, a comprehensive benchmark framework that systematically evaluates hallucination detection and mitigation across 72 configurations.
  arXiv  Detail & Related papers  (2026-05-04T10:43:27Z)
• The Semantic Illusion: Certified Limits of Embedding-Based Hallucination Detection in RAG Systems [0.0]
  We apply hallucination prediction to RAG detection, transforming scores into decision sets with finite-sample coverage guarantees.<n>We analyze this failure through the lens of distributional tails, showing that while NLI models achieve acceptable AUC (0.81), the "hardest" hallucinations are semantically indistinguishable from faithful responses.
  arXiv  Detail & Related papers  (2025-12-17T04:22:28Z)
• Detecting AI Hallucinations in Finance: An Information-Theoretic Method Cuts Hallucination Rate by 92% [4.693270291878929]
  Large language models (LLMs) produce fluent but unsupported answers - hallucinations.<n>We propose ECLIPSE, a framework that treats hallucination as a mismatch between a model's semantic entropy and the capacity of available evidence.
  arXiv  Detail & Related papers  (2025-12-02T05:25:48Z)
• CLUE: Non-parametric Verification from Experience via Hidden-State Clustering [64.50919789875233]
  We show that correctness of a solution is encoded as a geometrically separable signature within the trajectory of hidden activations.<n>ClUE consistently outperforms LLM-as-a-judge baselines and matches or exceeds modern confidence-based methods in reranking candidates.
  arXiv  Detail & Related papers  (2025-10-02T02:14:33Z)
• HIDE and Seek: Detecting Hallucinations in Language Models via Decoupled Representations [17.673293240849787]
  Contemporary Language Models (LMs) often generate content that is factually incorrect or unfaithful to the input context.<n>We propose a single-pass, training-free approach for effective Hallucination detectIon via Decoupled rEpresentations (HIDE)<n>Our results demonstrate that HIDE outperforms other single-pass methods in almost all settings.
  arXiv  Detail & Related papers  (2025-06-21T16:02:49Z)
• ShED-HD: A Shannon Entropy Distribution Framework for Lightweight Hallucination Detection on Edge Devices [2.193668764461763]
  Large Language Models (Ms.LS) have demonstrated impressive capabilities on a broad array of tasks, but their tendency to produce hallucinations.<n>We propose a novel hallucination detection framework that efficiently detects distinctive uncertainty patterns.
  arXiv  Detail & Related papers  (2025-03-23T23:47:26Z)
• A New Benchmark and Reverse Validation Method for Passage-level Hallucination Detection [63.56136319976554]
  Large Language Models (LLMs) generate hallucinations, which can cause significant damage when deployed for mission-critical tasks. We propose a self-check approach based on reverse validation to detect factual errors automatically in a zero-resource fashion. We empirically evaluate our method and existing zero-resource detection methods on two datasets.
  arXiv  Detail & Related papers  (2023-10-10T10:14:59Z)
• HyPoradise: An Open Baseline for Generative Speech Recognition with Large Language Models [81.56455625624041]
  We introduce the first open-source benchmark to utilize external large language models (LLMs) for ASR error correction. The proposed benchmark contains a novel dataset, HyPoradise (HP), encompassing more than 334,000 pairs of N-best hypotheses. LLMs with reasonable prompt and its generative capability can even correct those tokens that are missing in N-best list.
  arXiv  Detail & Related papers  (2023-09-27T14:44:10Z)
• A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation [76.34411067299331]
  Large language models often tend to 'hallucinate' which critically hampers their reliability. We propose an approach that actively detects and mitigates hallucinations during the generation process. We show that the proposed active detection and mitigation approach successfully reduces the hallucinations of the GPT-3.5 model from 47.5% to 14.5% on average.
  arXiv  Detail & Related papers  (2023-07-08T14:25:57Z)

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.