FACTUM: Mechanistic Detection of Citation Hallucination in Long-Form RAG

• URL: http://arxiv.org/abs/2601.05866v2
• Date: Fri, 16 Jan 2026 13:21:03 GMT
• Title: FACTUM: Mechanistic Detection of Citation Hallucination in Long-Form RAG
• Abstract summary: Retrieval-Augmented Generation (RAG) models are critically undermined by citation hallucinations.<n>We introduce FACTUM, a framework for testing Citation Trustworthiness via Underlying Mechanisms.<n>Our analysis reveals that correct citations are consistently marked by higher parametric force.<n>By capturing this complex interplay, FACTUM outperforms state-of-the-art baselines by up to 37.5% in AUC.
• Score: 29.07623863966094
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Retrieval-Augmented Generation (RAG) models are critically undermined by citation hallucinations, a deceptive failure where a model cites a source that fails to support its claim. While existing work attributes hallucination to a simple over-reliance on parametric knowledge, we reframe this failure as an evolving, scale-dependent coordination failure between the Attention (reading) and Feed-Forward Network (recalling) pathways. We introduce FACTUM (Framework for Attesting Citation Trustworthiness via Underlying Mechanisms), a framework of four mechanistic scores: Contextual Alignment (CAS), Attention Sink Usage (BAS), Parametric Force (PFS), and Pathway Alignment (PAS). Our analysis reveals that correct citations are consistently marked by higher parametric force (PFS) and greater use of the attention sink (BAS) for information synthesis. Crucially, we find that "one-size-fits-all" theories are insufficient as the signature of correctness evolves with scale: while the 3B model relies on high pathway alignment (PAS), our best-performing 8B detector identifies a shift toward a specialized strategy where pathways provide distinct, orthogonal information. By capturing this complex interplay, FACTUM outperforms state-of-the-art baselines by up to 37.5% in AUC. Our results demonstrate that high parametric force is constructive when successfully coordinated with the Attention pathway, paving the way for more nuanced and reliable RAG systems.

Related papers

• Synergistic Perception-Reasoning Governance: Grounding Medical MLLMs with Verifiable Anatomical Evidence [45.78409219344429]
  Multimodal large language models (MLLMs) show strong promise for clinical VQA and radiology report generation, yet inference-time hallucinations undermine trustworthy use.<n>We propose a holistic, training-free evidence-injection framework that systematically mitigates hallucinations through dual-side evidence injection.<n>We conduct systematic evaluations on 2 tasks and 5 datasets using textttLLaVA-1.5-7B, texttLLaVA-Med-1.5-7B, texttQwen3-VL-8B/32B, and textttInternVL
  arXiv  Detail & Related papers  (2026-06-30T08:07:25Z)
• R-APS: Compositional Reasoning and In-Context Meta-Learning for Constrained Design via Reflective Adversarial Pareto Search [18.8323743697237]
  Large language models (LLMs) are fluent on open-ended tasks, yet in agentic settings, fluency does not ensure reliable delivery.<n>We trace this gap to three coupled structural failures: errors propagate without localization, worst-case perturbations go unevaluated, and accumulated knowledge is never invalidated.<n>We argue these share a root cause: abductive, counterfactual, meta-inductive, corrective, and inductive reasoning pull a shared context in incompatible directions.
  arXiv  Detail & Related papers  (2026-06-03T12:45:39Z)
• The Devil is in the Spectrum: Mitigating Representation Collapse in LLMs via Topologically Regularized Side-Path [10.000577340093434]
  The Topologically Regularized Side-Path (TRSP) is a non-invasive architectural intervention that achieves spectral balance.<n>TRSP employs a parameter-free Triangular Box mechanism, scaled by a lightweight, length-aware gate, to regularize the token interaction topology.<n>Experiments show significant improvements across general capabilities and long-context benchmarks.
  arXiv  Detail & Related papers  (2026-06-02T09:11:50Z)
• 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)
• TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training [53.93696896939915]
  Training tool-use agents typically rely on Supervised Fine-Tuning (SFT) on successful trajectories and Reinforcement Learning (RL) on pass-rate-selected tasks.<n>We propose TopoCurate, an interaction-aware framework that projects multi-trial rollouts from the same task into a unified semantic quotient topology.<n>TopoCurate achieves consistent gains of 4.2% (SFT) and 6.9% (RL) over state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-03-02T10:38:54Z)
• MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systems [38.44649280816596]
  We propose MAS-FIRE, a systematic framework for fault injection and reliability evaluation of Multi-Agent Systems.<n>We define a taxonomy of 15 fault types covering intra-agent cognitive errors and inter-agent coordination failures.<n>Applying MAS-FIRE to three representative MAS architectures, we uncover a rich set of fault-tolerant behaviors.
  arXiv  Detail & Related papers  (2026-02-23T13:47:43Z)
• OMG-Agent: Toward Robust Missing Modality Generation with Decoupled Coarse-to-Fine Agentic Workflows [9.617220633655716]
  We present textbfunderlineOmni-textbfunderlineModality textbfunderlineGeneration Agent (textbfOMG-Agent)
  arXiv  Detail & Related papers  (2026-02-04T02:25:40Z)
• When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering [0.2796197251957245]
  We study whether synchronized iterative retrieval and reasoning can surpass an idealized static upper bound (Gold Context) RAG.<n>We benchmark eleven state-of-the-art LLMs under three regimes: (i) No Context, measuring reliance on parametric memory; (ii) Gold Context, where all evidence is supplied at once; and (iii) Iterative RAG, a training-free oracle controller that alternates retrieval, hypothesis refinement, and evidence-aware stopping.<n>Across models, Iterative RAG consistently outperforms Gold Context, with gains up to 25.6 percentage points, especially for non-reasoning
  arXiv  Detail & Related papers  (2026-01-27T17:35:05Z)
• Mechanisms of Prompt-Induced Hallucination in Vision-Language Models [58.991412160253276]
  We study the failure mode in a controlled object-counting setting, where the prompt overstates the number of objects in the image.<n>We identify a small set of attention heads whose ablation substantially reduces prompt-induced hallucinations (PIH) by at least 40% without additional training.<n>Our findings offer insights into the internal mechanisms driving prompt-induced hallucinations, revealing model-specific differences in how these behaviors are implemented.
  arXiv  Detail & Related papers  (2026-01-08T18:23:03Z)
• HalluZig: Hallucination Detection using Zigzag Persistence [0.1687274452793636]
  We introduce a new paradigm for hallucination detection by analyzing the dynamic topology of model's layer-wise attention.<n>Our core hypothesis is that factual and hallucinated generations exhibit distinct topological signatures.<n>We validate our framework, HalluZig, on multiple benchmarks, demonstrating that it outperforms strong baselines.
  arXiv  Detail & Related papers  (2026-01-04T14:55:43Z)
• Detecting Hallucinations in Graph Retrieval-Augmented Generation via Attention Patterns and Semantic Alignment [36.45654492179688]
  Graph-based Retrieval-Augmented Generation (GraphRAG) enhances Large Language Models (LLMs)<n>LLMs struggle to interpret the relational and topological information in inputs, resulting in hallucinations that are inconsistent with the retrieved knowledge.<n>We propose two lightweight interpretability metrics: Path Reliance Degree (PRD), which measures over-reliance on shortest-path triples, and Semantic Alignment Score (SAS), which assesses how well the model's internal representations align with the retrieved knowledge.
  arXiv  Detail & Related papers  (2025-12-09T21:52:50Z)
• Citation Failure: Definition, Analysis and Efficient Mitigation [56.09968229868067]
  Citations from LLM-based RAG systems are supposed to simplify response verification.<n>This does not hold for citation failure, when a model generates a helpful response, but fails to cite complete evidence.<n>We propose to disentangle this from response failure, where the response itself is flawed, and citing complete evidence is impossible.
  arXiv  Detail & Related papers  (2025-10-23T07:47:22Z)
• VAR: Visual Attention Reasoning via Structured Search and Backtracking [49.427842994857635]
  We introduce Visual Attention Reasoning, a framework that recasts grounded reasoning as a structured search.<n> VAR decomposes the reasoning process into two key stages: traceable evidence grounding and search-based chain-of-thought.<n>We show that our 7B model, VAR-7B, sets a new state-of-the-art on a comprehensive suite of hallucination and safety benchmarks.
  arXiv  Detail & Related papers  (2025-10-21T13:18:44Z)
• Continual Action Quality Assessment via Adaptive Manifold-Aligned Graph Regularization [53.82400605816587]
  Action Quality Assessment (AQA) quantifies human actions in videos, supporting applications in sports scoring, rehabilitation, and skill evaluation.<n>A major challenge lies in the non-stationary nature of quality distributions in real-world scenarios.<n>We introduce Continual AQA (CAQA), which equips with Continual Learning capabilities to handle evolving distributions.
  arXiv  Detail & Related papers  (2025-10-08T10:09:47Z)
• 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)
• Alignment Tipping Process: How Self-Evolution Pushes LLM Agents Off the Rails [103.05296856071931]
  We identify the Alignment Tipping Process (ATP), a critical post-deployment risk unique to self-evolving Large Language Model (LLM) agents.<n>ATP arises when continual interaction drives agents to abandon alignment constraints established during training in favor of reinforced, self-interested strategies.<n>Our experiments show that alignment benefits erode rapidly under self-evolution, with initially aligned models converging toward unaligned states.
  arXiv  Detail & Related papers  (2025-10-06T14:48:39Z)
• Beyond 'Aha!': Toward Systematic Meta-Abilities Alignment in Large Reasoning Models [86.88657425848547]
  Large reasoning models (LRMs) already possess a latent capacity for long chain-of-thought reasoning.<n>We explicitly align models with three meta-abilities: deduction, induction, and abduction, using automatically generated, self-verifiable tasks.<n>Our three stage-pipeline individual alignment, parameter-space merging, and domain-specific reinforcement learning, boosts performance by over 10% relative to instruction-tuned baselines.
  arXiv  Detail & Related papers  (2025-05-15T17:58:33Z)
• Retrieval is Not Enough: Enhancing RAG Reasoning through Test-Time Critique and Optimization [58.390885294401066]
  Retrieval-augmented generation (RAG) has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs)<n>RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions.<n>We propose AlignRAG, a novel iterative framework grounded in Critique-Driven Alignment (CDA)<n>We introduce AlignRAG-auto, an autonomous variant that dynamically terminates refinement, removing the need to pre-specify the number of critique iterations.
  arXiv  Detail & Related papers  (2025-04-21T04:56:47Z)
• Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling [78.78822033285938]
  Vision-Language Models (VLMs) excel at visual understanding but often suffer from visual hallucinations.<n>In this work, we introduce REVERSE, a unified framework that integrates hallucination-aware training with on-the-fly self-verification.
  arXiv  Detail & Related papers  (2025-04-17T17:59:22Z)
• Pairwise Matching of Intermediate Representations for Fine-grained Explainability [7.415710605852485]
  We propose a new explainability method (PAIR-X) that generates fine-grained, highly-localized pairwise visual explanations.<n>By improving interpretability, PAIR-X enables humans to better distinguish correct and incorrect matches.
  arXiv  Detail & Related papers  (2025-03-28T21:13:43Z)
• Hallucination, Monofacts, and Miscalibration: An Empirical Investigation [2.3278261859840104]
  Hallucinated facts in large language models obey a statistical lower bound determined by the monofact rate.<n>We present the first empirical investigation of this three-way relationship in classical n-gram models and fine-tuned encoder-decoder Transformers.
  arXiv  Detail & Related papers  (2025-02-11T18:46:00Z)
• Unfamiliar Finetuning Examples Control How Language Models Hallucinate [75.03210107477157]
  Large language models are known to hallucinate when faced with unfamiliar queries. We find that unfamiliar examples in the models' finetuning data are crucial in shaping these errors. Our work further investigates RL finetuning strategies for improving the factuality of long-form model generations.
  arXiv  Detail & Related papers  (2024-03-08T18:28:13Z)
• The Paradox of Motion: Evidence for Spurious Correlations in Skeleton-based Gait Recognition Models [4.089889918897877]
  This study challenges the prevailing assumption that vision-based gait recognition relies primarily on motion patterns. We show through a comparative analysis that removing height information leads to notable performance degradation. We propose a spatial transformer model processing individual poses, disregarding any temporal information, which achieves unreasonably good accuracy.
  arXiv  Detail & Related papers  (2024-02-13T09:33:12Z)
• A Multi-Level Attention Model for Evidence-Based Fact Checking [58.95413968110558]
  We present a simple model that can be trained on sequence structures. Results on a large-scale dataset for Fact Extraction and VERification show that our model outperforms the graph-based approaches.
  arXiv  Detail & Related papers  (2021-06-02T05:40:12Z)

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.