Large Language Model Hallucination

Tool-augmented large language model (LLM) agents fail in a way no tool-selection or tool-security method addresses: they call tools that do not exist and pass arguments no schema declares. Existing defenses either pick the right tool (selection) or constrain what an agent may do with real tools (gating), both of which presuppose the emitted call refers to a real tool at all. We show this is a structural blind spot: a hallucinated call is by construction not a decision any gate made, so no gate can reject it. This paper is primarily a measurement and benchmark study. We give a five-class taxonomy of tool hallucination (H1-H5) and, as a reference point, the Resolution Rung: a training-free, closed-world resolver (registry membership plus a signature check) whose interest is where it must sit, not what it computes. We prove hallucination defense must precede any causal gate, and characterize the one irreducible residue (borrowed arguments schema-indistinguishable from a valid call). Across ten hosted models under two invocation surfaces we measure 322 genuine hallucinations; fabricated-tool calls concentrate on the unconstrained raw-JSON surface (34 vs. 3), and model scale does not help (a 675B model matches a 7-8B one). We then extend to the Model Context Protocol, where merging several servers into one namespace creates hallucination surfaces a single registry cannot express (a second taxonomy, M1-M5); on the live MCP surface we measure 154 hallucinations, including from frontier models that were clean on the single-registry surface, because collisions and shadowing are structural to the merge. We release the versioned Hallucinated-Tools Benchmark (HTB) so any resolver is comparable across submissions.

Multimodal large language models (MLLMs) have made strong progress on visual question answering and image captioning, yet they still produce fluent claims about objects, attributes, or relations that are not grounded in the image. Many remedies either modify decoding at test time, which adds latency, or fine tune with preferences such as DPO variants, which teach which answer is preferred but not when the model's own answer is unreliable. We argue that calibrated self assessment is the missing signal. We introduce Savor, a training framework that (i) augments the output schema with token and answer confidence, (ii) optimises the policy with a Group Relative Policy Optimisation (GRPO) objective that penalises calibration error and poor abstention decisions, and (iii) uses the learned confidence at inference time to revisit visual evidence only when the model is uncertain. Experiments on POPE, HallusionBench, AMBER and MMHal-Bench across two recent backbones (InternVL3-8B and Qwen3-VL-8B) show that Savor reduces hallucination while preserving general capability on MME and MMBench, with lower Expected Calibration Error than DPO and decoding baselines.

Hallucination remains a critical challenge in large language models (LLMs), where early factual errors compound through autoregressive generation in a snowballing effect that neither post-hoc correction nor weight-level intervention can effectively preempt. We propose DescaPE (DEcoding Signal Control Against Path Error-snowballing), a decoding framework that leverages internal model signals to suppress hallucination-prone trajectories at inference time. Through sliding-window MLP ablation, we identify a factual-salient layer span within LLMs whose derived signal is selectively elevated for factual tokens and exhibits anomalous spikes at hallucination-prone steps. We train a lightweight probe to approximate this signal from a single forward pass and integrate it into candidate scoring to penalize high-risk continuations while rewarding factually grounded ones. Experiments across five factuality benchmarks on three LLMs demonstrate that DescaPE achieves factuality improvements over decoding-time baselines in multiple settings, while incurring only 1.10x latency overhead in our efficiency evaluation. Our code is available at https://github.com/hayeonggg/DESCAPE.

Large audio-language models (LALMs) can hallucinate audio objects, answering "yes" to absent sound events, thus undermining reliability in audio question answering. We propose Token-Adaptive Decoding (TAD), a training-free strategy for hallucination mitigation that grounds the initial yes/no decision by contrasting logits under real audio with a matched silent reference. TAD introduces a token-adaptive, confidence-guided gate that is decision-critical at the first decoding step and class-conditional on affirmative tokens, using the audio-silent margin to avoid overcorrection when evidence is weak or already sufficient. Experiments on AudioCaps-Hallucination show that, relative to Audio-Aware Decoding (AAD), a contrastive baseline with fixed contrast strength, TAD improves F1 for Qwen2 by 0.059 to 0.117 across Popular, Adversarial, and Random splits, and for Gemma by 0.025 to 0.064, while on Clotho-AQA it raises F1 from 0.810 to 0.816 on Qwen2 and remains comparable to AAD on Gemma.

Recent inference-time hallucination mitigation methods for large vision-language models (LVLMs) report strong gains on hallucination benchmarks. However, it remains unclear whether lower hallucination scores reflect improved multimodal grounding or more conservative generation. We evaluate six mitigation methods across three LVLMs and four benchmarks, including hallucination-focused evaluation and the diverse capability benchmark MMStar. Our analysis reveals two consistent patterns. First, hallucination reduction is often coupled with reduced informativeness: methods that lower hallucination rates also reduce object recall, visual coverage, or response detailedness. Second, improvements on hallucination benchmarks do not reliably transfer to broader multimodal capabilities, with methods showing inconsistent or degraded performance on fine-grained perception and reasoning tasks. Our findings suggest that current evaluation protocols may overestimate progress by rewarding conservative generation. We argue that hallucination mitigation should be evaluated as a faithfulness--informativeness--capability trade-off rather than through hallucination scores alone.

Diffusion-based Large Vision-Language Models (dLVLMs) have recently emerged as a compelling alternative to autoregressive (AR) LVLMs, offering advantages in parallel decoding, bidirectional context, and controllable generation. Despite rapid progress, their reliability properties remain largely uncharacterized. We present the first systematic reliability evaluation of hallucination and bias in dLVLMs, benchmarking six diffusion models against competitive AR baselines across four dimensions. Our key findings are: (1) dLVLMs reverse the yes-bias of AR models in binary visual queries; (2) they achieve competitive hallucination rates yet exhibit degraded linguistic quality; (3) they collapse to near-zero accuracy on underrepresented racial groups with opposite-polarity gender bias; and (4) they exhibit accuracy collapse in multiple-choice settings when the correct option is shorter than its distractors, associated with a length prior that emerges at the first denoising step. Tokens committed at late denoising steps with low confidence further correlate with hallucinated content, pointing to a mechanistic signal unique to diffusion generation. These patterns vary across model families, suggesting reliability is shaped by the generative paradigm together with training data.

Both privacy and factual accuracy are paramount in high-stakes domains like healthcare. Concerningly, we uncover and investigate a privacy-hallucination tradeoff in differentially private (DP) language models. First, we empirically show that models pre-trained or fine-tuned with DP tend to produce more hallucinations than non-DP counterparts, with increased severity as the privacy budget grows stricter. Second, we investigate model properties driving this tradeoff, demonstrating that DP mechanisms flatten output distributions, potentially redistributing probability mass toward factually incorrect alternatives. Third, through experiments where we control fact frequency in training data, we characterize how information frequency can reduce hallucination risks in DP models. Overall, our findings underscore the need for more nuanced privacy-preserving interventions that offer rigorous privacy guarantees without compromising factual accuracy.

Existing research on object hallucination in multimodal large language models (MLLMs) predominantly attributes the problem to language priors such as over-reliance on textual co-occurrence statistics. We challenge this view by presenting quantitative evidence for a complementary, under-explored cause: visual-origin hallucination, where hallucinations arise from incorrect visual feature extraction and misalignment between image and text embeddings. Through cosine similarity analysis and Smooth Grad-CAM entropy measurements, we show that hallucinated samples exhibit systematically lower image-text similarity (average 0.158 vs. -0.122) and inverted attention patterns, where attention is dispersed when the target object is present but wrongly concentrated when it is absent. Guided by this diagnosis, we propose Adversarial Contrastive Fine-Tuning (ACFT). ACFT uses an Adversarial Hallucination Attribute Flipping (AHAF) procedure, involving minimal, targeted adversarial perturbations that flip an image's hallucination attribute, to construct perfectly aligned positive-negative pairs, which are then used for contrastive fine-tuning. AHAF simultaneously serves as a diagnostic probe, revealing that MLLM visual representations lie dangerously close to hallucination decision boundaries. Requiring only 0.9% of the COCO dataset and adding zero inference overhead, ACFT achieves state-of-the-art performance on POPE, MME, and four description-level hallucination benchmarks across LLaVA, MiniGPT-4, and Qwen2.5-VL. Code is available at https://github.com/zxp555/ACFT_MM

Large Vision-Language Models (LVLMs) achieve strong performance across many multimodal tasks; however, they often exploit spurious object-background correlations, resulting in predictions driven by contextual shortcuts rather than object-relevant visual evidence. Despite growing interest in hallucination and robustness evaluation, existing benchmarks provide limited control over whether model predictions are grounded in the target object or induced by correlated background cues. In this work, we introduce PURGE (\underline{P}artition-aware \underline{U}nlearning for \underline{R}emoving spurious-correlation \underline{G}enerated \underline{E}rrors), a framework for constructing, benchmarking, and mitigating spurious-correlation-induced failures in LVLMs. The framework consists of: -- (1) Structured dataset construction wherein we develop three complementary structured data construction strategies that partition examples by object-relevant evidence and spurious background cues, enabling controlled diagnosis of shortcut reliance; and -- (2) Partition-aware unlearning, which uses these partitions to selectively remove spurious object-background associations while preserving object-based reasoning. We evaluate the \algo~framework across multiple LVLMs, including LLaVA-1.6-7B, Qwen3-VL-8B-Instruct, and Qwen3.5-9B, together with CLIP as a vision-language encoder, on a diverse suite of benchmarks, including CHAIR, POPE, Causal-HalBench, MM-SpuBench, AMBER, MMHal, and Waterbirds. Our results show that PURGE consistently reduces hallucinations and spurious-correlation-driven errors while maintaining or improving overall performance in most evaluated settings, providing both a reusable evaluation protocol and an effective mitigation framework for more reliable LVLMs.

Large Vision-Language Models (LVLMs) are prone to hallucinations: they fluently describe objects, attributes, and scenes that are not in the image. We connect part of this failure to a measurable property of their representations, feature instability, where mild semantics-preserving perturbations of the input cause large changes in the learned embeddings; hallucination rates rise together with this variability. Existing stability-motivated remedies are explicit, in the sense that they intervene at inference time through latent steering or constrained decoding, and pay for it on every query. We propose implicit stabilization instead: perturbation-invariance is built into the model weights during fine-tuning, and nothing extra runs at deployment. Our framework, INFUSE, first stabilizes visual and textual representations around perturbation-averaged and ground-truth anchors, then aligns the stabilized representations across modalities with bidirectional contrastive objectives. We prove that the anchor's root-mean-square deviation from the perturbation-mean representation shrinks at rate $1/\sqrt{K}$ in the number of views, and that under a Lipschitz decoder, this bounds how much any perturbation can change the model's hallucination behavior. On LLaVA-1.5, LLaVA-1.6, and Qwen3-VL-8B-Instruct, INFUSE reduces AMBER CHAIR by 46-63% relative to each base model, improves ObjHal, MMHal, HallusionBench, and POPE, and preserves VQA-v2 and TextVQA, all with no inference-time overhead.

Uncertainty quantification (UQ) methods are widely used for hallucination detection in large language models (LLMs) in closed-book settings where ground-truth evidence is unavailable at inference time. Prior work has proposed combining UQ signals via learned ensembles, but empirical investigations into the robustness of these ensembles are limited. We study a supervised ensembling framework that trains a classifier over heterogeneous UQ-based scorer outputs on a small, domain-specific dataset of labeled LLM responses, then applies it to out-of-sample hallucination classification without retrieval, tools, or reference documents. Across four LLMs, nine datasets, and three generation regimes (short-form QA, long-form generation, and code generation), we provide a systematic robustness analysis along three axes: sample efficiency, in-domain dataset transfer, and generation regime dependence. We find that supervised ensembles outperform the best individual scorer in 30 of 32 settings, with gains realized from as few as 100 labeled instances. Ensembles retain most of their advantage in cases of in-domain transfer under distribution shift, outperforming the best non-ensemble scorer in 23 of 28 transfer settings. Sampling-based black-box ensembles are nearly as effective as full ensembles, while single-generation white-box ensembles offer limited benefit.

Multimodal large language models (MLLMs) have made rapid progress, yet they still exhibit object hallucination, generating plausible but incorrect descriptions that are inconsistent with the visual input. Direct Preference Optimization (DPO) mitigates this by training models to prefer non-hallucinated responses over hallucinated ones, and recent efforts further enrich the preference data with relevant context. However, it remains unclear whether DPO actually leverages such context. To investigate this, we propose Contextual Preference Gain (CPG), a simple metric that measures how much a model's preference strengthens when relevant context is provided. We find that higher CPG consistently corresponds to lower hallucination, yet standard DPO and its variants exhibit only limited CPG, indicating that they underutilize contextual information and thus remain prone to hallucination. To address this, we propose Context-Calibrated DPO (C$^2$-DPO), which directly maximizes CPG while preserving the original preference ordering. Across multiple benchmarks, C$^2$-DPO substantially reduces hallucination without compromising general reasoning, relatively reducing the Object HalBench hallucination rate of Qwen2-VL-Instruct-2B by 36%. Code is available at https://github.com/mlvlab/C2-DPO

Recent advances in large vision-language models (LVLMs) have enabled powerful multimodal reasoning by integrating visual encoders with large language models (LLMs). However, their reliability is frequently undermined by hallucinations, where generated text inaccurately describes the visual input. Although fine-tuning can mitigate this problem, it is computationally expensive and requires large, curated datasets, making training-free alternatives attractive. Among these, model editing is more promising than decoding-based approaches: decoding methods adapt outputs per input but introduce computational overhead and instability, whereas model editing modifies internal representations offline, providing a more efficient and stable solution. However, existing model-editing techniques typically rely on a single global subspace to correct hallucinations, treating all test samples identically and failing to capture diverse hallucination modes across inputs. To address this limitation, we propose a training-free hallucination mitigation framework for dynamic, per-instance suppression at test time. Our method first constructs a set of Disentangled Hallucination Subspaces, each isolating a distinct hallucination mode. During inference, the model adaptively calculates weights reflecting each input's relationship to these subspaces, guiding a dynamically combined projection that selectively suppresses the most probable hallucination directions while preserving image-grounded semantics. Extensive experiments across multiple vision-language benchmarks and LVLM families demonstrate consistent improvements, highlighting the robustness, generalizability, and efficiency of our approach.

Current LLM safety guardrails face a fundamental tension: fine-tuning distorts pre-trained representations while generative judges incur prohibitive inference costs. We ask a complementary question: how far can safety be achieved through pure geometric reasoning over frozen representations, and where does it fail? We present HoloAegis, a minimally parametric topological inference framework that decouples representation from reasoning: an un-fine-tuned encoder maps text to the unit sphere S^{d-1}, and all decisions reduce to Gibbs-Boltzmann free-energy differences over pre-computed anchor centroids. We contribute a boundary-mapping study rather than a leaderboard claim. On a frozen three-benchmark protocol, HoloAegis (3.2 MB) statistically matches WildGuard-7B (14 GB) on toxicity (0.96 vs. 0.96), exceeds it on harmful behaviors (0.99 vs. 0.79), and cedes oversafety detection (0.62 vs. 0.98) -- while ShieldGemma-2B fails on indirect harms (0.34). These failure modes are complementary and mechanistically traceable: potential-difference scoring senses manifold clustering, whereas policy-conditioned LLM judging requires explicit taxonomy matching. We restate our Topological Boundary Stability conjecture in ratio form and validate it via reference-set bootstrap: anchor banks reduce score variance 4-15x and boundary displacement to approximately 0.44 + 0.23 sqrt(k/K) of the full-space estimator. Per-domain analysis further reveals that geometric separability tracks within-domain semantic homogeneity. Our results chart where geometric guardrails substitute for, and where they must defer to, LLM judges.

Vision-language models (VLMs) excel at open-ended captioning and visual QA but often describe objects, attributes, or relations absent from the image, a phenomenon known as object hallucination. We propose a {training-free, post-hoc representation editing technique} that operates in the representation space of the language backbone. The method performs a lightweight, one-time offline calibration on a modest paired dataset to estimate the required covariance structures, using only forward passes and empirical second-order statistics with no gradient updates or fine-tuning, after which the correction is absorbed directly into the model's existing weights. By modeling hidden states as a superposition of truthful and hallucination-associated components, we derive a Wiener-type estimator whose optimal gains are given in closed form from the covariances of paired truthful and hallucinated representations. An eigendecomposition yields mode-wise attenuation that respects a stability criterion, i.e., the filter responds continuously to estimation noise. The correction is applied once to the feed-forward output projections of selected deeper layers, at inference time, the model runs unchanged and at the same speed. Experiments on LLaVA-1.5, MiniGPT-4, Gemma3, and mPLUG-Owl2 demonstrate consistent reductions in object hallucination on CHAIR, POPE, and MME while maintaining caption fluency and overall response quality. We further demonstrate the generality of our approach on the TempCompass video understanding benchmark and on discrete diffusion language models for grounded dialogue, showing that representation filtering reduces hallucinations even in temporal video reasoning and multi-step, sequence-wide denoising settings.

Large language models (LLMs) can generate fluent and useful responses but remain prone to hallucinations. We introduce Prompt Embedding Probes (PEP), a white-box method for answer-level hallucination detection from the hidden states of a frozen LLM. PEP extends standard linear probes by augmenting the input with a small number of learnable prompt embeddings. We evaluate PEP on TriviaQA, GSM8K, and MedQA using Qwen3 models at multiple scales. PEP improves hidden-state-based detection over standard linear probes in the main in-distribution setting. We further evaluate PEP for pre-generation prediction, cross-model transfer, and out-of-distribution generalization. PEP remains effective in the pre-generation and cross-model settings, whereas robust cross-dataset transfer remains difficult. These results show that prompt-based adaptation can strengthen hidden-state probing while keeping the backbone frozen and adding only a small number of trainable parameters.

Large reasoning models (LRMs) are prone to hallucination, which undermines their reliability and poses challenges for safe deployment. Hallucinations in LRMs arise from two distinct failure sources: reasoning hallucination, where flawed inference steps propagate to an incorrect conclusion, and knowledge hallucination, where the model lacks the requisite factual knowledge to answer the query. To address reasoning hallucination, we propose REIN, an alignment framework that trains LRMs to produce a structured reasoning sequence, \texttt{<think>} $$\rightarrow \texttt{<reflection>} $$\rightarrow $\texttt{<answer>}$, enabling explicit self-reflection before committing to a final answer. To address knowledge hallucination, REIN introduces a reward mechanism that encourages explicit abstention (e.g., "I don't know") when none of the sampled reasoning chains yields a correct answer, allowing the model to refrain from unsupported predictions. Extensive evaluations on mathematical and commonsense reasoning benchmarks show that REIN consistently improves selective accuracy, reduces incorrect-but-self-endorsed responses, and maintains high coverage compared with competitive baselines. Notably, REIN achieves these gains within a single forward pass, without requiring process supervision, inference-time controllers, external search, or multi-round critiques. Experiments on multiple backbones show that REIN reduces the hallucination proxy by $58\sim72\%$ relative to the base models while maintaining $86\sim91\%$ average coverage, and improves selective accuracy on attempted questions by $6.6\sim14.2\%$.

Large language models (LLMs) are often used in conjunction with external knowledge sources to improve their factual accuracy and decrease hallucinations, through methods such as Retrieval-Augmented Generation (RAG). However, these systems remain susceptible to intrinsic hallucinations, where the model generates unfaithful or fabricated information that is not supported by the retrieved evidence. We propose a novel framework to assess model robustness against this phenomenon by stress-testing using natural, semantically equivalent variations of a user query found via adversarial optimization methods. We apply our framework, which enforces strict semantic equivalence constraints and an intrinsic hallucination objective, to a range of adversarial attack techniques across white-box, gray-box, and black-box adversarial settings. Evaluating these attacks on 5 open-source and 5 closed-source generator models across 3 datasets, we demonstrate that even state-of-the-art models are highly susceptible to meaning-preserving perturbations, which significantly degrade contextual faithfulness (by up to 50% for GPT-5-mini). Our findings indicate that faithful use of in-context evidence remains fragile even in state-of-the-art LLMs, motivating architectures and training objectives that enforce robust grounding independent of surface query form. Code is available at: https://github.com/atriviveksharma/intrinsic_hall

Hallucination remains a critical challenge for developing trustworthy Multimodal Large Language Models (MLLMs). While existing benchmarks mainly focus on entity, attribute, and relation hallucinations, knowledge-related failures are often investigated separately, lacking a unified evaluation framework across different hallucination dimensions. To overcome this, we propose \textbf{KnowHal}, a benchmark that explicitly incorporates knowledge hallucination into multimodal hallucination evaluation spanning four dimensions: entity, attribute, relation, and knowledge. KnowHal constructs paired positive and negative questions over shared images and entities, enabling controlled comparisons among perceptual errors, knowledge-related errors, and false-premise acceptance. The benchmark contains 1,800 samples across 10 domains and 50 categories, constructed through a semi-automated pipeline combining LLM assistance, CLIP-based filtering, and human verification. We evaluate 14 representative MLLMs on KnowHal and conduct extensive analyses. Results show that the knowledge dimension consistently presents the greatest challenge for nearly all evaluated models, while most models exhibit substantial performance degradation on negative questions, revealing limited robustness to false premises. By unifying four hallucination dimensions with paired question design, KnowHal addresses an important gap in existing evaluation frameworks and enables a more comprehensive assessment of hallucinations in MLLMs.

Despite remarkable progress in visual understanding, Multimodal Large Language Models (MLLMs) remain prone to hallucinations when reasoning about spatial relationships, often producing judgments that contradict the true 3D structure of the scene. Though several existing works have proposed to mitigate hallucinations, our analysis indicates that they show limited effectiveness in spatial reasoning, as they fail to bridge the fundamental gap between 2D visual representations and 3D spatial reality. Based on this finding, we define hallucinations arising from insufficient spatial structure modeling as spatial reasoning hallucination, a subcategory of relation hallucination that existing mitigation methods fail to address. We further identify three typical scenarios where such hallucinations frequently occur: perspective effects, object orientation, and viewpoint changes. To this end, we propose Geo3R, a training-free, plug-and-play framework that incorporates geometric evidence and structured 3D reasoning to mitigate spatial reasoning hallucination. Experiments on three benchmarks, covering 18 tasks across all three scenarios, show that Geo3R substantially reduces spatial reasoning hallucination across diverse MLLMs without additional training, outperforming existing models and methods.

With the rapid advancement of large language models (LLMs), modern systems not only possess strong foundational capabilities and extensive knowledge, but can also solve complex problems via long, multi-step reasoning. However, as reasoning traces become longer, LLMs may produce a substantial amount of hallucinated content during the reasoning process, which is often difficult to detect. In this work, we conduct a fine-grained analysis of hallucinations arising in LLM reasoning and find that the reasoning traces are particularly prone to Context-Sensitive Factual Hallucinations: cases where the model actually has the relevant knowledge, yet makes factual errors due to contextual interference during reasoning. To address this issue, we propose Step-level Self-Consistency Group Relative Policy Optimization (SSC-GRPO), which assigns step-level rewards to reasoning traces by computing self-consistency scores of individual steps across multiple rollouts. Compared with prior methods, SSC-GRPO achieves state-of-the-art performance on both mathematical reasoning benchmarks and hallucination leaderboards. Our results offer a new perspective for detecting and mitigating hallucinations in the reasoning process of large language models.

Large vision-language models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding. However, they remain prone to hallucinations, generating responses that are inconsistent with the visual evidence. Existing mitigation methods largely address language-prior bias or cross-modal imbalance, while progressive visual degradation across perception and memory remains underexplored. In this work, we propose Saliency-Driven Perceptual Realignment (SDPR), a training-free framework that mitigates the degradation of visual awareness throughout inference. Specifically, we first introduce saliency-driven attention redistribution to release attention hijacked by non-semantic sink tokens, thereby recovering critical visual evidence. Second, we identify spatial distortion in the KV cache and propose saliency-driven cache alignment to preserve query-relevant visual features during generation. Finally, we introduce prior-constrained contrastive decoding to penalize unfaithful predictions induced by dominant language priors. Our proposed SDPR is robust against hallucinations due to its holistic alignment of visual awareness across the entire generative trajectory. Extensive experiments across diverse LVLM architectures show that SDPR outperforms state-of-the-art methods on both hallucination and general-purpose benchmarks, requiring no additional training and incurring minimal runtime overhead. The code is available \href{https://github.com/PengSyuChen/SDPR}{\color{blue}{here}}.

Hallucinations and artificial text in LLM-generated outputs often appear as distributional deviations between prompt and response hidden-state distributions. Since prompts or retrieved contexts typically serve as reference samples and responses as query samples, with major differences in length, these asymmetries motivate the use of change test statistics that treat the two samples differently. We consider an asymmetric two-sample test ASK-NN based on the directed k-nearest-neighbor graph. Our statistic counts reference points whose nearest neighbor in the pooled sample is also a reference point. Under the permutation null, it admits an exact finite-sample conditional mean and variance; we further establish asymptotic normality and consistency under fixed alternatives. ASK-NN is computationally effective and easy to implement. Empirically, it is competitive with kernel and graph-based baselines on synthetic benchmarks, artificial-text detection, and LLM hallucination detection from token-level hidden states.

While recent advances in 3D generation have enabled impressive visual synthesis, existing methods often rely on 2D diffusion supervision without explicit mechanisms for geometric consistency, leading to spatial hallucinations such as duplicated structures and misaligned geometry. These issues become more severe in 4D generation, where maintaining consistency across viewpoints and temporal evolution introduces additional challenges, including jitter, identity flicker, and structural drift. We present \textbf{Hallo4D}, a unified and model-agnostic framework for mitigating spatiotemporal hallucinations in 3D and 4D content generation. Hallo4D introduces a generation-detection-correction paradigm that leverages large multimodal language models (LMMs) to identify and summarize spatial and temporal inconsistencies from multi-view and multi-frame renderings. These insights guide a consensus-driven image-space consistency optimization, where an LMM-based selector evaluates candidate corrections through multi-model voting, without requiring retraining or architectural modifications. To further improve temporal consistency and optimization efficiency, Hallo4D incorporates motion-aware keyframe sampling, LMM-guided initialization, and appearance alignment. We additionally introduce exposure-aware optimization and visibility pruning to enhance robustness under challenging viewpoints. Extensive experiments demonstrate that Hallo4D consistently outperforms strong baselines across diverse 3D and 4D generation settings, providing a scalable and generalizable solution for consistency-aware content generation.

Tool access alone does not make LLM empirical reasoning governable: accepted outputs need not descend from attested evidence, and accepted deductions need not hold up under formal scrutiny. We present EG-VAR (Evidence-Grounded Verified Agentic Reasoning), a Lean 4-based tool-calling architecture in which the Lean kernel is the sole minter of Verified claims via tool-attestation axioms and declared source lifts. Every verified output structurally descends from an attested tool call (Thm. 3.1) and a kernel-checked chain of valid inference (Thm. 3.2); residual outputs are honest Abstain with a replayable audit trail. On a subcollection of TableBench numerical reasoning (n=120), EG-VAR attains 120/120 versus a 95% same-tool baseline; on counterfactual stress tests (5 domains x 2 models), EG-VAR stays 100% source-faithful while same-tool drops to 80-90% (no-tool 50-80%). With the LLM as deployment-time formalizer, residual semantic-formalization error is 3.3% on Sonnet and 1.7% on Opus. We position EG-VAR as a technical-governance interface for high-stakes empirical claims: a formal sidecar makes the target proposition, source scope, evidence boundary, proof obligation, and abstention condition auditable, eliminating unsupported Verified outputs today while turning formalization errors, lift and source-authority disputes, ambiguities, and abstentions into explicit audit targets. Over time, typed sidecars in datasets, APIs, public records, and AI-generated documents can amortize this formalization burden into reusable infrastructure.

Recent advances in equipping Large Language Models (LLMs) with search tools and outcome-reward reinforcement learning (RL) have achieved new state-of-the-art results on open-domain QA tasks. However, we argue that current training paradigms harbor a critical vulnerability: they predominantly reward correct answers but fail to penalize fabricated ones when retrieval fails, thereby implicitly exacerbating hallucinations. To address this, we propose Abstention-Aware Reinforcement Learning (AWA-RL), which dynamically shapes the abstention reward utilizing the model's query-specific prior capabilities and continuous on-policy training observations. We also introduce a novel metric, RA-F1, to measure the capability-reliability trade-off. Compared to non-abstaining baselines, AWA-RL boosts absolute precision by up to 10.3% and overall RA-F1 by 2.9%, with only marginal sacrifice in raw accuracy. These results confirm that AWA-RL successfully yields highly capable and reliable search agents. The code, data, and model weights are publicly available at https://github.com/zfj1998/AWA-RL.

Large reasoning models (LRMs) improve language model capabilities by generating explicit thinking traces before final answers. In factuality-oriented question answering (QA), such thinking often improves overall performance by helping the model recover relevant knowledge and refine its answers. However, we find that this benefit is not uniform at the instance level: explicit thinking can also overturn correct non-thinking answers and lead to factual drift. We refer to this failure mode as \emph{thinking-induced hallucination}. To explain this phenomenon, we formulate explicit thinking in factuality QA as a thinking residual over the model's direct-answer tendency, which can either recover missing knowledge or introduce unsupported associations. Based on this formulation, we propose MARGO, \underline{\textit{M}}ixed-Mode \underline{\textit{A}}dvantage \underline{\textit{R}}egularization for \underline{\textit{G}}rounded \underline{\textit{O}}ptimization, a reinforcement learning framework that uses non-thinking rollouts as same-model references in advantage estimation. By constructing mixed-mode rollout groups with both thinking and non-thinking trajectories, MARGO evaluates whether explicit thinking adds factual value beyond direct answering, thereby suppressing hallucination-prone thinking while preserving beneficial thinking behaviors. Experiments across multiple factuality-oriented QA benchmarks demonstrate that MARGO improves factual reliability over strong baselines, while evaluations on mathematical benchmarks show that it preserves general reasoning ability.

Large Vision-Language Models (LVLMs) have achieved remarkable progress in visual understanding tasks such as image captioning and visual question answering. However, they remain susceptible to hallucinations, generating content that is inconsistent with the actual visual input. Existing methods primarily intervene at the decoding stage, while overlooking a critical source of hallucinations: irrelevant or noisy visual tokens that mislead the decoding process. To address this issue, we propose SeeMe, a training-free framework that introduces the concept of feature engineering from traditional machine learning into LVLMs. SeeMe restructures visual tokens through a three-stage token engineering process to suppress hallucination sources while preserving informative visual evidence. Experiments on MME, POPE, and AMBER benchmarks across four LVLMs demonstrate that SeeMe consistently reduces hallucinations and improves output consistency, providing a novel perspective for mitigating hallucinations in LVLMs.

Finetuning a language model on documents that are explicitly annotated as fictional results in a model that still actually believes the documents' core claims, an effect known as Negation Neglect. In our evaluations, models trained on documents prefixed and suffixed with such annotations correctly identify the relevant claims as fictional only about 9% of the time. To address this, we introduce Goggles, a learned module that intervenes on the finetuning gradient rather than the data. During supervised finetuning, a Goggles module edits the gradients an LLM LoRA receives, imparting a chosen epistemic frame (the stance the model takes toward the nature of what it reads) to whatever the documents teach. A Goggles instance is trained once for a given base model, frame, and LoRA configuration, then applied frozen to documents it was never trained on. Trained through Goggles on those same documents, now carrying no fictional annotation, the model flags the content as fictional roughly 91% of the time, while preserving capability (GPQA and TruthfulQA match or exceed baseline). The same architecture supports other frames: a Goggles instance can be trained to treat documents as "part of an AI safety evaluation by Redwood Research" rather than simply as fiction. The imparted frame persists under continued finetuning that pushes back toward the claim, where prior interventions revert. Goggles suggests a path toward training language models on known-misaligned data without absorbing the behaviors that data demonstrates.

Large language models (LLMs) are increasingly applied to resume optimization for applicant tracking systems, introducing hallucination failures distinct from general text generation: anachronistic technology injection, cross-domain terminology contamination, structural mutation, and content fabrication. We present Grounded Optimization, a five-layer framework combining temporal context validation, deterministic contamination detection, structural invariant enforcement, prompt-level grounding, and an evaluator agent. In ablation experiments across three LLMs, four temperature settings, and six layer configurations on 25 synthetic resumes spanning 14 industries, undefended baselines produce 2.48-5.36 detected hallucinations per resume. Among detectors independent of the active defenses, temporal hallucinations are reduced by 50-95% across all conditions; overall detected hallucination rate falls to 0.04-0.24. Prompt-level grounding alone achieves zero detected hallucinations at low temperature with a capable instruction-following model; higher temperatures and weaker models reveal the need for the deterministic layers as a complement. We release the contamination taxonomy, evaluation code, and raw data.

Large language models often produce hallucinated answers that violate prompt-level constraints. A key diagnostic question is whether these failures reflect missing knowledge, or whether the model has the relevant information but follows the wrong inference path. We study this phenomenon as inference misalignment: a mismatch between the answer supported by the prompt and the answer favored by statistically salient latent associations. We formalize this view with a latent key-task model, in which pretraining-frequency imbalance can cause a shortcut path to dominate the constraint-sensitive path and induce positive inference loss. The framework predicts two failure modes: task-retrieval bias in entity disambiguation and key-selection bias in action choice. We introduce TrapQA, a controlled diagnostic testbed with two components. ScientistQA tests disambiguation among similar scientists with supplementary factual probes, while Real-Life Constrained QA tests everyday constraint following under salient shortcuts. Our results show that hallucination can arise from biased latent inference rather than absent knowledge alone.

Multimodal large language models (MLLMs) show strong promise for clinical VQA and radiology report generation, yet inference-time hallucinations still undermine trustworthy use: models can produce fluent conclusions that conflict with imaging evidence. Existing mitigation strategies typically rely on additional training, external retrieval/knowledge bases, or multi-stage post-hoc verification, which increases cost and pipeline complexity and often generalizes poorly across models and tasks.To address this, we propose a holistic, training-free evidence-injection framework that systematically mitigates hallucinations through dual-side evidence injection. By leveraging ROI priors acquired using MedSAM in our implementation, we recalibrate the visual perception trajectory via ROI-guided activation modulation while anchoring the textual reasoning trajectory by mapping anatomical coordinates into discrete semantic tokens as verifiable external memory. Then we introduce a task-aware dynamic router to select modality-specific interventions based on task semantics, balancing perceptual grounding and linguistic fluency. We conduct systematic evaluations on 2 tasks and 5 datasets using \texttt{LLaVA-1.5-7B}, \texttt{LLaVA-Med-1.5-7B}, \texttt{Qwen3-VL-8B/32B}, and \texttt{InternVL-3.5-8B/38B}. Controlled ablations and visualizations further validate the framework, which consistently outperforms baselines across medical benchmarks, improving close-ended accuracy by up to $\sim\mathbf{6}\%\uparrow$ and reducing open-ended hallucinations by $\sim\mathbf{35}\%\downarrow$. The code has been made available on GitHub: \href{https://github.com/Henry991115/SPRG}{\textcolor{blue}{https://github.com/Henry991115/SPRG}}.

Multimodal Large Language Models (MLLMs) are critically hampered by hallucination, generating content inconsistent with the provided image. In this paper, we identify an internal signature of hallucination: progressive degradation of text-to-image cross-attention during generation, leading to specific failure patterns like unfocused or biased attention. Existing mitigation strategies are largely outcome-driven and do not explicitly target this failure mode. To address this problem, we propose ADAPT (Attention Dynamics Alignment with Preference Tuning), an attention-based framework that intervenes directly on text-to-image cross-attention dynamics. We propose ADAPT with three key contributions: a cross-attention visual anchor refined from early decoding to provide stable spatial grounding, an attention-supervised inference mechanism that detects and corrects attention drift online, and a Visual Attention Guidance DPO that aligns preferences toward visually grounded responses. Experiments show that each component of ADAPT contributes to hallucination reduction, and the full framework achieves new best results across multiple hallucination benchmarks, reducing hallucination rates by 40%-60% across mainstream backbones while preserving general multimodal capabilities. Our work provides an attention-based perspective on mitigating hallucinations by exploring the model's internal text-to-image cross-attention behaviors. Code is available at https://github.com/yao-ustc/ADAPT

Large Vision-Language Models (LVLMs) excel at multimodal tasks but remain prone to object hallucinations. Prior training-free remedies often uniformly strengthen visual signals, which may also amplify irrelevant regions and introduce spurious evidence, harming fluency. We propose Context-aware Attention Intervention (CAI), a training-free inference-time mechanism that enforces a see only when needed principle via two-axis selectivity: where to look and when to intervene. At each decoding step, CAI derives token-specific visual relevance from early-layer representations to localize semantically aligned regions, and applies a conservative, entropy- and depth-gated attention tilt only for uncertainty-spiking tokens in deeper layers where visual grounding degrades, leaving confident tokens and irrelevant regions largely unchanged. This targeted intervention strengthens visual grounding while preserving linguistic fluency, and it yields consistent improvements even without contrastive decoding, which remains optional as an auxiliary bias-suppression module. Extensive experiments across multiple LVLM backbones and benchmarks show that CAI achieves state-of-the-art hallucination mitigation, and our analysis characterizes CAI as a KL-minimal attention reweighting with bounded interference under inactive gates or small tilts. Code is available at https://github.com/Iris1946/CAI.

Multimodal Large Language Models (MLLMs) are prone to hallucination as their generation preferences are insufficiently calibrated to visual evidence, causing them to fall back on linguistic priors, rather than faithful grounding. In this work, we start from an empirical observation: when query-relevant visual evidence is explicitly strengthened using the model's own attention, generation becomes more accurate, suggesting that many failures do not arise solely from missing perception, but from an insufficient tendency to trust the evidence the model has already attended to. Motivated by this finding, we propose Oriented Pickup Preference Optimization (\texttt{OPPO}), an evidence-aware alignment objective that learns preferences over the strength of visual evidence, rather than only response quality. Concretely, \texttt{OPPO} contrasts the same faithful response under stronger, anchored, weaker-evidence views, turning naive visual preference into ordered visual-evidence alignment. We further combine this objective with fine-grained span-level and token-level regularization to stabilize the training. Besides, we provide a theoretical analysis showing that ordered evidence margins induce a positive lower bound on local visual sensitivity. Extensive evaluations across hallucination and general-purpose benchmarks demonstrate that \texttt{OPPO} consistently outperforms baseline methods.

Despite the impressive capabilities of Large Vision-Language Models (LVLMs), they remain susceptible to hallucination, generating content inconsistent with the input image. Recent studies attribute this to the dominance of language priors over visual inputs and employ contrastive decoding methods to mitigate this dominance, but the mechanistic origin remains unexplored. We investigate the information flow through each transformer layer and find that attention modules consistently aggregate visual evidence, while FFN modules at critical layers act as the source of language priors. These priors can override visual evidence, causing correct predictions in intermediate layers to drift toward incorrect outputs. Based on this insight, we propose FADE (FFN Attenuation for DEcoding), a training-free method that attenuates FFN outputs to reduce language-prior dominance. Evaluations on POPE, CHAIR, and MME benchmarks across LLaVA-1.5, mPLUG-Owl2, and InstructBLIP show that FADE effectively mitigates hallucinations while preserving inference efficiency.

Modern generative world models render increasingly realistic action-controllable futures, yet they frequently hallucinate: rollouts remain visually fluent while drifting from the ground-truth dynamics. We hypothesize that hallucination concentrates in low-coverage regions of the state-action space, where lightweight data-centric signals can both detect it and guide mitigation. To test this, we introduce MMBench2, a 427-hour, 210-task dataset for visual world modeling with ground-truth actions, rewards, and live simulators, and train a 350M-parameter world model on it. We identify three distinct hallucination modes: perceptual, action-marginalized, and scene-diverging -- each anchored to a different stage of the pipeline, and develop three signals that accurately predict where the model will fail. To close coverage gaps at training time, we develop a coverage-aware sampling technique; to close them online, our hallucination predictors serve as curiosity rewards for targeted data collection, yielding a data-efficient finetuning recipe that adapts the pretrained world model to entirely unseen environments with as few as 50 real environment trajectories. Overall, our findings reveal that hallucination in world models is inherently a data coverage issue, and that the same signals used to detect it can also be used for mitigation. An interactive web version of our paper is available at https://www.nicklashansen.com/mmbench2

Transcatheter Aortic Valve Replacement (TAVR) planning requires meticulous multimodal reasoning. However, adapting Multimodal Large Language Models (MLLMs) to this high-stakes domain is severely impeded by diagnostic hallucinations, where generated text lacks anatomical grounding. To address this, TAVR-VLM is introduced: a novel framework featuring Risk-Conditioned Causal Grounding Attention (R-CGA) that instantiates a model-internal ``Risk $\rightarrow$ Region $\rightarrow$ Word'' structural grounding pathway. R-CGA compresses multimodal inputs into a causal risk bottleneck, purifying dense visual features into a global risk mask. During autoregressive generation, a support-projected causal consistency objective constrains token-level grounding within the risk-defined support mask. Evaluated on $\text{M}^3\text{TAVR}$, a comprehensive 1,482-patient cohort, TAVR-VLM establishes a new state-of-the-art. It achieves an AUROC of 0.896, boosts CIDEr to 0.936, and drastically reduces the hallucination rate to 8.1%, thereby improving interpretability for evidence-based surgical AI.

Machine learning models for remote sensing are trained and deployed on a static set of modalities. However, as we equip newer satellites with novel sensors and retire old ones, practitioners may wish to deploy an existing model on a substitution, superset, or subset of modalities with minimal retraining given data availability or practical computational constraints. We study the setting of updating existing models to changing modalities and identify three main scenarios: Modality Transfer (substitution), Addition (superset), and Peeking (subset). We propose DeluluNet, an architecture with modular components for all three changing modality scenarios. DeluluNet is trained end-to-end, learning a multi-modal model from a unimodal teacher and unlabeled multimodal data via modality hallucination--predicting missing modality representations from those that are present. As a result, DeluluNet can keep predicting even when input modalities change, providing a practical alternative to re-labeling and re-training in a changing world.

Vision-language models (VLMs) often generate fluent but visually unsupported descriptions, especially by mentioning objects absent from the image. We propose QK Product Steering, a data-free, training-free, and zero-inference-cost weight edit for reducing object hallucination. The method directly edits the per-head query-key product, the operator that produces pre-softmax attention logits, by suppressing a small number of dominant singular modes in selected middle layers. The edited product is then mapped back to the query weights through a closed-form query-only update while keeping shared key weights fixed, making the edit compatible with grouped-query attention. We further decompose the QK product into symmetric and antisymmetric components to distinguish mutual content-similarity patterns from directional attention patterns. Across three GQA-based VLMs, QK Product Steering achieves an average relative CHAIR$_s$ reduction of $4.0\%$, while matched random-mode controls show negligible change. Interpretability ablations show that the hallucination signal is specific to dominant QK modes and is primarily localized to the symmetric mutual-attention channel. Overall, QK Product Steering offers a simple alternative to decoding-time mitigation, requiring no additional data, fine-tuning, or inference-time overhead while largely preserving general multimodal capability.

While Multimodal Retrieval-Augmented Generation (M-RAG) enhances Large Vision-Language Models, it remains highly susceptible to cross-modal hallucinations, causal fabrications, and sycophancy. Furthermore, existing mitigation pipelines often face an intervention paradox: static rules tend to unnecessarily disrupt accurate generations, whereas leaving the multi-modal reasoning completely unguided allows existing mismatches to cascade into severe logical fabrications. To quantify and mitigate these hallucinations, we propose a Multi-Agent system, MODE-RAG, driven by Variational Free Energy (VFE) and internal attention states to dynamically gate interventions. High-risk queries are routed to five stage-specific agents, integrating Monte Carlo Tree Search (MCTS) for rigorous causal derivation and logit perturbations to penalize sycophancy. Dedicated Correction and Overseer agents ensure formatting stability and perform post-hoc factual verification. To objectively evaluate our approach, we introduce ModeVent, a challenging subset derived from the MultiVent dataset. Extensive experiments indicate that our system effectively reduces hallucination rates and logical fabrication, significantly improving the robustness of M-RAG systems.

Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided evidence. Reinforcement learning (RL) is a promising direction for hallucination mitigation, but response-level faithfulness rewards suffer from a granularity mismatch: localized hallucinations can cause supported content to receive spurious penalties. Although recent work introduces fine-grained feedback such as claim-level verification and token-level rewards, unbalanced credit assignment can still induce length, verbosity, or optimization-noise biases. We propose BALTO, a Balanced Token-level Policy Optimization framework for hallucination mitigation. BALTO extracts checkable factual claims, verifies them against the reference context, and projects claim-level judgments to token-level labels. A balanced token-level credit assignment mechanism is introduced into the framework. This design redistributes probability mass from unsupported content toward faithful content, rather than suppressing the entire response. We systematically analyze the limitations of response-level rewards from a theoretical standpoint, and prove BALTO's advantages in training stability and optimization efficiency for hallucination mitigation. Experiments on ConFiQA, RAGTruth, and FinLLM-Eval show that BALTO achieves the highest faithfulness across all six model--benchmark settings and consistently outperforms existing post-training baselines in Q-Score, demonstrating a stronger faithfulness--informativeness trade-off.

Large Language Models (LLMs) are increasingly deployed in healthcare settings, yet their tendency to hallucinate poses risks when clinical decisions are involved. This study examine whether LLMs recommend recently banned or withdrawn pharmaceuticals when answering clinical questions and tests an agent-based method for reducing such errors. We developed a five-agent "Trust but Verify" system using a single LLM backbone. To measure regulatory knowledge obsolescence, we created an adversarial dataset of 103 clinical MCQs where historically correct answers now refer to banned substances. This scale ensures statistical significance across various therapeutic classes. We evaluated three open-access model families (GPT-OSS, Llama-3, Falcon-3) under vanilla and agentic conditions. Performance was measured via pointwise score, label accuracy, Hallucination Error Rate (HER), and Component Fidelity (CF) score. We also observed clinical safety regression in proprietary models. In default configurations, all models showed high hallucination rates, consistently selecting banned drugs that matched training data patterns. Our proposed agentic architecture reduced HER by approximately 53% across models. Pointwise scores shifted from -0.25 (unsafe recommendation) toward 0.0 (appropriate refusal). The safety audit intercepted dangerous outputs even when models' parametric knowledge favored the banned substance. The proposed multi-agent framework offers a model-agnostic method for enforcing regulatory compliance that prioritizes patient safety over fluent text generation. Our work demonstrates a practical approach for deploying autonomous AI systems in safety-critical healthcare settings. It shows how real-time regulatory data can be integrated into LLM pipelines to support clinical decision-making.

The rapid updates of online news make tracking event developments challenging, highlighting the need for timeline summarization (TLS). Hallucinations, where LLM-generated content deviates from source news, still remain a critical issue in LLM-based TLS and are not well studied in existing works. To bridge this gap, we identify two primary types of hallucinations: unfaithful content during news summarization and information omission in date-event summarization. Then, we propose NTS-CoT, a novel framework that leverages Chain-of-Thought (CoT) reasoning to mitigate hallucinations in TLS. The framework consists of three key modules: i) Element-CoT to capture essential news elements for faithful summarization, ii) Date Selection to combine temporal saliency and event prominence for timestamp selection, and iii) Causal-CoT to infer causal relationships and reduce omissions in date-event summarization. Extensive experiments, including quantitative analysis on three TLS benchmarks and human evaluation, demonstrate that NTS-CoT outperforms state-of-the-art baselines, effectively mitigating hallucinations and improving LLM-based TLS performance. Our source code is available at https://anonymous.4open.science/r/NTS-CoT .

Large language models (LLMs) often hallucinate by generating factually incorrect or unfaithful content, posing significant risks to their safe use. Detecting such hallucinations is particularly challenging under the zero-source constraint, where no model internals or external references are available, and detection must rely solely on the textual query-answer pair. In this paper, we propose Human-like Criteria Probing for Hallucination Detection (HCPD), a paradigm that emulates the multi-faceted reasoning of human evaluators. Its core is a Human-like Criteria Probing (HCP) mechanism, in which a LLM agent adaptively decomposes its judgment into a weighted set of interpretable criteria and aggregates criterion-specific scores into a final truthfulness measure. To achieve this adaptive capability, we introduce a reward-based alignment scheme using only weak supervision from semantic consistency. At inference, we employ a multi-sampling aggregation strategy to ensure robust decisions while preserving full interpretability. We further provide theoretical analysis supporting the reliability of our approach. Extensive experiments show that HCPD consistently outperforms state-of-the-art baselines, offering an effective and explainable solution for zero-source hallucination detection. Code is available at https://github.com/TRISKEL10N/HCPD.

Large language models (LLMs) are increasingly used to access organisational documentation, including standard operating procedures (SOPs), HR policies and institutional guidelines. However, retrieval-augmented generation (RAG) systems that rely on free-form rewriting can introduce hallucinations and unstable trade-offs between completeness and conciseness, particularly in safety- and compliance-critical settings. Objectives: To evaluate extraction as a hallucination-resistant alternative to rewriting-based RAG and compare strategies that balance precision, recall and safety across document types and model scales. Methods: We compare multiple prompting strategies, including line-number-based source selection, extraction of relevant guideline sentences with explicit safety annotations, and a multi-stage pipeline that refines draft answers using supporting evidence from source guidelines. Experiments are conducted on documents of varying length and structure, including local NHS acute care and oncology guidelines and UK-wide NICE guidelines, using both frontier-scale and locally deployable models. Performance is assessed using automatic metrics and human expert evaluation of relevance and completeness. Results: Line-number selection achieves the strongest results, outperforming direct copying and safety-focused strategies across both large and small models while maintaining high term recall (up to 95%) and close alignment with source text. Safety-oriented approaches improve precision but introduce systematic omissions, while multi-stage filtering further amplifies this trade-off. Performance varies with document structure: line-based extraction excels in protocol-like content, whereas alternative strategies perform better on more verbose documents (up to 97% term recall).

Hallucinations -- factually incorrect or unverifiable outputs -- remain one of the most challenging limitations of Large Language Models (LLMs), especially in knowledge-intensive tasks. One proposed explanation is internal knowledge conflicts arising from fixed, outdated training data. This paper investigates whether internal representations linked to knowledge conflicts correlate with hallucination behaviors in LLMs. Using probing techniques inspired by two prior works, we analyzed activations from hidden, attention, and MLP layers, as well as output logits, across predefined tasks. We probed LLaMA-3-8B on hallucination detection benchmarks and Falcon-7B on a knowledge conflict dataset. Our findings show that, although conceptually related, hallucination activation patterns cannot be fully reduced to or explained by knowledge conflict representations. Nonetheless, probing proves a robust tool across multiple languages and activation types, supporting its role in improving LLM interpretability. This work advances the broader understanding of hallucinations in LLMs and underscores the value of fine-grained analysis of their internal behavior.

Large language models (LLMs) can generate factually inconsistent claims, motivating accurate and scalable hallucination detectors. Prior work largely enlarges training sets via synthesis or new annotations, introducing increasing cost and potential bias while underusing the consistency implied by semantically equivalent paraphrases. We propose Consistency-Constrained Hallucination Detector (CCHD), which formulates training as a constrained optimization problem. The standard cross-entropy on original document-claim pairs is complemented by (i) paraphrase-consistency constraints bounding divergence across paraphrased views, and (ii) label-preservation constraints tying paraphrases to ground truth. We solve the problem by gradient descent-ascent over model parameters and per-view Lagrange multipliers, adding only a few scalar dual variables and no inference-time overhead. With DeBERTa and Flan-T5 backbones, CCHD consistently outperforms strong baselines (FactCG, MiniCheck, and AlignScore) on standard factuality benchmarks, demonstrating its superiority on hallucination detection.

Large language models (LLMs) frequently generate hallucinations, which are unsupported by a source document. To avoid costly LLM-as-evaluator pipelines and the heavy annotation demands of existing classifiers, we propose CPIL (Cross Paraphrastic Invariance Learning), a two-stage Siamese framework that maximizes the utility of existing labeled data. Concretely, CPIL constructs informative training pairs by: (i) generating paraphrastic views of each document-claim example as positives, and explicitly aligning their representations to enforce invariance to surface form; and (ii) mining same-document, opposite-label pairs as hard negatives to sharpen document-sensitive decision boundaries. Then CPIL conduct a two-stage model training: Stage 1 performs contrastive pretraining to learn a paraphrase-invariant, grounding-aware embedding space; and Stage 2 attaches a lightweight classifier for binary groundedness. On the LLM-AggreFact benchmark (11 tasks), CPIL surpasses strong baselines concerning F1 scores with only ~1% labeled data, showing its prediction superiority and label efficiency.

Large language models (LLMs) have achieved remarkable progress in open-ended text generation, yet they remain prone to hallucinating incorrect or unsupported content, which undermines their reliability. This issue is exacerbated in long-form generation due to hallucination snowballing, a phenomenon where early errors propagate and compound into subsequent outputs. To address this challenge, we propose a novel inference-time hallucination mitigation framework, named Segment-wise HAllucination Rejection Sampling (SHARS), which uses an arbitrary hallucination detector to identify and reject hallucinated segments during generation and resample until faithful content is produced. By retaining only confident information and building subsequent generations upon it, the framework mitigates hallucination accumulation and enhances factual consistency. To instantiate this framework, we adopt semantic uncertainty as the detector and introduce several vital modifications to address its limitations and better adapt it to long-form text. Our method enables models to self-correct hallucinations without requiring external resources such as web search or knowledge bases, while remaining compatible with them for future extensions. Empirical evaluations on standardized hallucination benchmarks demonstrate that our method substantially reduces hallucinations in long-form generation while preserving or even improving the informativeness of generation. Code is available at: https://github.com/TreeLLi/hallucination-rejection-sampling.

Large vision language models (LVLMs) have made rapid advancements and are deployed across various applications, yet hallucinations remain a major challenge. Activation steering is appealing due to its minimal training overhead and controllability at inference time. However, we found that during autoregressive decoding, visual conditioning affects token prediction sparsely and locally across decoding steps, and many existing methods that average image-versus-no-image differences over the entire sequence dilute these critical signals, yielding low signal-to-noise ratio steering directions. Additionally, many existing methods apply a fixed steering strength, which misallocates the intervention budget, over-perturbs non-critical tokens, and can cause instability. To address these limitations, we propose Token-Level Visual-Sensitivity Steering (TLVS) for hallucination mitigation. Our approach first extracts token-level steering vectors and refines them, and then applies fine-grained, visual-sensitivity-adaptive steering only where it matters. This lightweight, plug-and-play mechanism requires only minimal training for calibration and can be applied across diverse vision-language models. It modulates the steering strength at each decoding step, selectively suppressing hallucination-prone spans while preserving evidence-grounded content. We evaluate TLVS on several benchmarks, including POPE, AMBER, CHAIR (COCO), MMHal, and HallusionBench, demonstrating consistent improvements over previous steering methods.

Small open-source code models that power IDE autocomplete still emit hallucinated Fill-in-the-Middle (FIM) completions: syntactically natural calls to methods, parameters, variables, and imports that do not exist in the surrounding project. Existing mitigations either require per-language execution sandboxes that do not apply at mid-keystroke or preference-optimisation pipelines that need large human-labelled corpora. We propose an execution-free alternative: use frontier code models to synthesise plausible-but-wrong completions as hard negatives, then leverage the contrast between these synthetic hallucinations and the ground-truth developer edit as a supervised fine-tuning signal. Our pipeline scrapes multilingual FIM contexts from public GitHub across eight languages and asks a panel of three frontier generators to produce one hard negative per context for each of four hallucination types drawn from the Delulu taxonomy, a Docker-verified multilingual FIM hallucination benchmark, yielding a paired chosen/rejected dataset. Fine-tuning Qwen2.5-Coder-7B-Instruct on a 100K-row curated subset lifts Delulu exact match by +18.8 points and edit similarity by +0.22 on every language and every type, while also improving every HumanEval-Infilling split and every SAFIM subset. The same recipe at 3B lifts Delulu by +12.8 EM with a small, characterised general-FIM trade-off. Five-axis ablations (size, type mix, language coverage, base-model family, and a difficulty-aware fool rate) plus a head-to-head SFT vs. DPO/ORPO comparison map which design choices drive the gain. We release the full pipeline source code -- generation, fool-rate LLM judging, curation, and the FIM fine-tuning recipe -- so that the experiments in this paper can be reproduced end-to end on any permissively licensed corpus.

Hallucination in Large Language Models (LLMs), characterized by the generation of content inconsistent with contextual facts or logical constraints -- remains a persistent challenge for reliable deployment. In this work, we address this issue through a geometric framework rooted in the linear representation hypothesis. We propose that hallucinations manifest as orthogonal noise relative to the semantic manifold of the residual stream. Specifically, we hypothesize that while attention heads ideally propagate information congruent with the context subspace, hallucinations arise when specific heads introduce components orthogonal to this subspace, disrupting the coherence of the latent representation. Based on this formulation, we introduce Dynamic Contextual Orthogonalization (DCO), an inference-time intervention method. DCO utilizes the input residual stream as a dynamic context anchor to perform orthogonal decomposition on attention head outputs. To distinguish between context-aligned semantic updates and divergent noise, DCO employs a layer-wise Z-score suppression mechanism that selectively attenuates outlier orthogonal components based on statistical distributions. Evaluations on Llama-3-8B and 70B across benchmarks such as XSum, NQ-Swap, and IFEval demonstrate that DCO achieves superior contextual faithfulness compared to state-of-the-art intervention baselines. Furthermore, DCO maintains high performance on knowledge-intensive tasks like TriviaQA and TruthfulQA, effectively mitigating the trade-off between hallucination suppression and parametric knowledge retention often observed in existing methods. Our findings validate the geometric interpretation of hallucinations and establish DCO as a computationally efficient approach for enforcing manifold alignment.Our code is available at https://github.com/Harry-Miral/DCO

Existing hallucination taxonomies classify LLM errors by what is wrong with the output -- memorised misconceptions, reasoning failures, fluent fabrications. These taxonomies are useful for diagnosis but cannot answer a different question: which uncertainty scorer would have caught this error? We propose a complementary taxonomy that classifies errors by their detectability signature -- the signal a scorer family would read. The DECK taxonomy is a 2x2 partition along inter-sample consistency and token-level confidence into four behavioural regimes (Drift, Entrenched, Confabulation, Knotted), each mapping to a specific scorer family (or families) that can detect it: black-box consistency scorers have signal in D and C, white-box token-probability scorers have signal in K and C, and only an LLM-as-a-Judge with independent pretraining can detect E. Cell membership is operationalised by a Youden's J optimal split on each scorer axis. Across three models and four datasets we validate the taxonomy two ways: by analysing scorer-pair disagreement, and by checking that external labels (SelfAware unanswerable, HaluEval adversarial, PopQA entity popularity) land in the predicted DECK cells, with model-scale and content-specific secondary-cell refinements. We further identify a universal blind spot of output-level UQ: on knowledge-gap inputs where the generator emits confident, repeatable fabrications, every output-level family collapses by construction. A linear probe on Llama-3-8B's hidden states also collapses to chance, giving preliminary evidence that the failure may persist at the activation level; richer internal-state methods (UQ heads, information-theoretic estimators) remain to be tested.

Hallucinations in medical large language models (LLMs) pose serious risks for clinical decision support, particularly when models must reason over complex electronic health records (EHRs). However, existing benchmarks often lack a realistic clinical context and provide limited insight into how hallucinations can be mitigated in practice. We introduce Med-HEAL, a framework for systematically identifying, analyzing, and mitigating hallucinations in medical LLMs using clinically grounded data. Building on the EHRNoteQA benchmark derived from MIMIC-IV discharge summaries, we construct a hallucination dataset by evaluating BioMistral-7B on open-ended clinical question answering tasks. Model outputs are labeled through a dual evaluation pipeline that combines LLM-as-a-Judge assessment (GPT-4o) with human auditing by medical student reviewers, producing correctness judgments and annotations of reasoning errors via a custom web-based evaluation system. We then leverage this dataset to investigate mitigation strategies: a self-critique pipeline, in which the test model reviews its own answers to detect potential errors and regenerates responses for flagged cases, and retrieval-augmented in-context learning (RA-ICL), which exposes the model to hallucinated and corrected examples. Experiments across five open-source LLMs-BioMistral, Llama-3.1, DeepSeek, Qwen2.5, and Qwen3, show that the self-critique strategy improves accuracy for three of five models (p < 0.05) without requiring parameter updates. Med-HEAL provides both a reusable hallucination dataset and a practical framework for studying and mitigating hallucinations in medical LLMs, supporting safer deployment of AI systems in clinical environments. Our code and data are publicly available at https://github.com/yimingliao-blad/med-heal.git.

MLLMs frequently hallucinate objects inconsistent with visual inputs. This issue is typically attributed to the over-reliance on language priors, which can override the visual context. Recent training-free decoding strategies address this by penalizing language priors. However, these methods overlook the dual nature of language priors, where they can be both helpful and harmful depending on the alignment with visual evidence. In particular, blindly suppressing language priors often disrupts the model's semantic manifold, leading to performance degradation, a phenomenon we term Manifold Departure. To address this, we propose Manifold-Guided Adaptive Projection (MGAP), a geometry-aware, training-free decoding method that mitigates hallucinations while preserving representation structure. MGAP first constructs a language-prior subspace from blind hidden states via SVD. During decoding, MGAP projects each multimodal hidden state onto this subspace and applies a consistency-aware gate to adaptively attenuate only the projected prior component, yielding a subspace-selective update that largely preserves the orthogonal semantic components. Extensive experiments on POPE and CHAIR show that MGAP outperforms prior decoding baselines, achieving stronger hallucination suppression without sacrificing coherence.

Large language models (LLMs) have seen widespread adoption across various domains, yet their reliability is frequently undermined by hallucinations - responses that are plausible-sounding but factually incorrect. In high-stakes domains, these errors can reduce trust and introduce real-world risk. To address this challenge, we present a parameter-efficient approach that uses soft prompts to mitigate hallucinated content and promote responsible abstention in generative question-answering (QA) tasks. Our method, called Responsible Contrastive Soft Prompting (RCSP), uses a composite loss to train soft prompts that balance three goals: suppressing hallucinatory content, encouraging abstention under uncertainty, and preserving or improving factual recall. To achieve these goals, we incorporate contrastive loss, curriculum learning, and KL regularization into our training mechanism. We evaluate our approach on five diverse generative QA datasets using an LLM-as-a-Judge framework. Experimental results on the Gemma 3 (12B) and Llama 3.1 (8B) backbones demonstrate that RCSP effectively balances factual recall with hallucination suppression and abstention, yielding a generally superior F-score over standard reasoning and instruction-based prompting baselines. Notably, these improvements are achieved by training only a fraction of the parameters required by other tuning techniques. Our results demonstrate that soft prompts provide a modular and computationally efficient path toward improving LLM reliability.

Large Language Models (LLMs) have achieved strong performance across diverse natural language tasks, yet their outputs often suffer from hallucinations -- content that is misaligned with factual information. In this work, we conduct a comprehensive layer-wise analysis of the decoding process and reveal that hallucinations tend to originate from deeper decoder layers. To address this issue, we introduce \textbf{DeLask} (\textbf{De}coder \textbf{La}yer \textbf{Sk}ipping), a novel decoding framework that dynamically skips layers prone to producing hallucinations. DeLask leverages the theoretical insight that the forward computation of an $L$-layer Transformer is conditionally equivalent to $L$ steps of gradient descent. We define a \emph{driftance value} by computing the cosine similarity between gradients derived from consecutive decoder steps, identifying problematic layers when the descent direction reverses. Rather than discarding such layers entirely, DeLask partially aggregates their hidden states with preceding layers, thereby preserving consistency while suppressing erroneous signals. Extensive experiments across diverse LLMs and benchmarks demonstrate that DeLask consistently mitigates hallucinations and enhances overall reliability, providing a lightweight and generalizable decoding framework for improving the robustness of large-scale language models.

Large language models hallucinate even when the answer lies within their parameters. While inference-time scaling can surface this latent knowledge, the most effective methods require supervision: a trained verifier or reward model. We ask what can be done with only a base language model: which intrinsic signal best identifies correct outputs, and how should it be decoded? We cast this as a score~$\times$~decoder grid pairing four scores (perplexity, contrastive, power-distribution likelihood, and self-verification) with three decoding families (optimization, sampling, consensus), and evaluate every cell on MATH500 with the base and instruction-tuned Qwen3-1.7B. While self-verification, which prompts the model to judge its own answer and is sharpened by a training-free virtual-thinking prefix, works well in most settings, no score has a fixed quality: its value depends on the decoder that consumes it and on model capability. When no supervision is available, the score and the decoding family must be chosen together.

Multimodal large language models (MLLMs) demonstrate remarkable visual understanding, yet their reliability in interactive settings is severely undermined by hallucination snowballing: a phenomenon where initial errors amplify across conversational turns, leading to a collapse in coherence. This failure reveals a fundamental vulnerability where models progressively neglect visual grounding in favor of over-relying on polluted textual history. Existing benchmarks are predominantly confined to single-turn VQA, which fail to capture the complex dynamics of error propagation in long-horizon interactions. To address this, we introduce MM-Snowball, the first benchmark for fine-grained diagnosis of hallucination snowballing within dialogues. Extensive evaluation shows that our benchmark poses a significant challenge even to advanced MLLMs and reveals the inefficacy of existing mitigation methods designed for single-turn VQA. To counteract this degradation, we propose Conflict-Aware Visual Rectification (CAVR). This training-free method mitigates snowballing through a synergistic dual-mechanism that refreshes visual grounding at the representation level and rectifies output distributions at the logit level, effectively re-anchoring the model to visual facts. Experiments demonstrate that CAVR achieves state-of-the-art performance, offering a promising path toward more reliable interactive AI. Data and code are available at: https://frenkie-chiang.github.io/MM-Snowball