Large language models (LLMs) integrated with retrieval-augmented generation (RAG) have improved factuality by grounding outputs in external evidence. However, they remain susceptible to unfaithful generation, where outputs contradict retrieved context despite its relevance and accuracy. Existing approaches aiming to improve faithfulness primarily focus on enhancing the utilization of external context, but often overlook the persistent influence of internal parametric knowledge during generation. In this work, we investigate the internal mechanisms behind unfaithful generation and identify a subset of mid-to-deep feed-forward networks (FFNs) that are disproportionately activated in such cases. Building on this insight, we propose Parametric Knowledge Muting through FFNSuppression (ParamMute), a framework that improves contextual faithfulness by suppressing the activation of unfaithfulness-associated FFNs and calibrating the model toward retrieved knowledge. To evaluate our approach, we introduce CoFaithfulQA, a benchmark specifically designed to evaluate faithfulness in scenarios where internal knowledge conflicts with accurate external evidence. Experimental results show that ParamMute significantly enhances faithfulness across both CoFaithfulQA and the established ConFiQA benchmark, achieving substantial reductions in reliance on parametric memory. These findings underscore the importance of mitigating internal knowledge dominance and provide a new direction for improving LLM trustworthiness in RAG.

Vision-Language models (VLMs) show impressive abilities to answer questions on visual inputs (e.g., counting objects in an image), yet demonstrate higher accuracies when performing an analogous task on text (e.g., counting words in a text). We investigate this accuracy gap by identifying and comparing the circuits--the task-specific computational sub-graphs--in different modalities. We show that while circuits are largely disjoint between modalities, they implement relatively similar functionalities: the differences lie primarily in processing modality-specific data positions (an image or a text sequence). Zooming in on the image data representations, we observe they become aligned with the higher-performing analogous textual representations only towards later layers, too late in processing to effectively influence subsequent positions. To overcome this, we patch the representations of visual data tokens from later layers back into earlier layers. In experiments with multiple tasks and models, this simple intervention closes a third of the performance gap between the modalities, on average.

Interpreting deep neural networks through concept-based explanations offers a bridge between low-level features and high-level human-understandable semantics. However, existing automatic concept discovery methods often fail to align these extracted concepts with the model's true decision-making process, thereby compromising explanation faithfulness. In this work, we propose FACE (Faithful Automatic Concept Extraction), a novel framework that augments Non-negative Matrix Factorization (NMF) with a Kullback-Leibler (KL) divergence regularization term to ensure alignment between the model's original and concept-based predictions. Unlike prior methods that operate solely on encoder activations, FACE incorporates classifier supervision during concept learning, enforcing predictive consistency and enabling faithful explanations. We provide theoretical guarantees showing that minimizing the KL divergence bounds the deviation in predictive distributions, thereby promoting faithful local linearity in the learned concept space. Systematic evaluations on ImageNet, COCO, and CelebA datasets demonstrate that FACE outperforms existing methods across faithfulness and sparsity metrics.

This paper investigates the structure of linear operators introduced in Hernandez et al. [2023] that decode specific relational facts in transformer language models. We extend their single-relation findings to a collection of relations and systematically chart their organization. We show that such collections of relation decoders can be highly compressed by simple order-3 tensor networks without significant loss in decoding accuracy. To explain this surprising redundancy, we develop a cross-evaluation protocol, in which we apply each linear decoder operator to the subjects of every other relation. Our results reveal that these linear maps do not encode distinct relations, but extract recurring, coarse-grained semantic properties (e.g., country of capital city and country of food are both in the country-of-X property). This property-centric structure clarifies both the operators' compressibility and highlights why they generalize only to new relations that are semantically close. Our findings thus interpret linear relational decoding in transformer language models as primarily property-based, rather than relation-specific.1

Large Reasoning Models (LRMs) have significantly enhanced their capabilities in complex problem-solving by introducing a thinking draft that enables multi-path Chain-of-Thought explorations before producing final answers. Ensuring the faithfulness of these intermediate reasoning processes is crucial for reliable monitoring, interpretation, and effective control. In this paper, we propose a systematic counterfactual intervention framework to rigorously evaluate .

Knowledge graph-based retrieval-augmented generation seeks to mitigate hallucinations in Large Language Models (LLMs) caused by insufficient or outdated knowledge. However, existing methods often fail to fully exploit the prior knowledge embedded in knowledge graphs (KGs), particularly their structural information and explicit or implicit constraints. The former can enhance the faithfulness of LLMs' reasoning, while the latter can improve the reliability of response generations. Motivated by these, we propose a trustworthy reasoning framework, termed Deliberation over Priors (\texttt{DP}), which sufficiently utilizes the priors contained in KGs. Specifically, \texttt{DP} adopts a progressive knowledge distillation strategy that integrates structural priors into LLMs through a combination of supervised fine-tuning and Kahneman-Tversky Optimization, thereby improving the faithfulness of relation path generation. Furthermore, our framework employs a reasoning-introspection strategy, which guides LLMs to perform refined reasoning verification based on extracted constraint priors, ensuring the reliability of response generation. Extensive experiments on three benchmark datasets demonstrate that \texttt{DP} achieves new state-of-the-art performance, especially a H@1 improvement of 13% on the ComplexWebQuestions dataset, and generates highly trustworthy responses. We also conduct various analyses to verify its flexibility and practicality.

Large language models (LLMs) integrated with retrieval-augmented generation (RAG) have improved factuality by grounding outputs in external evidence. However, they remain susceptible to unfaithful generation, where outputs contradict retrieved context despite its relevance and accuracy. Existing approaches aiming to improve faithfulness primarily focus on enhancing the utilization of external context, but often overlook the persistent influence of internal parametric knowledge during generation. In this work, we investigate the internal mechanisms behind unfaithful generation and identify a subset of mid-to-deep feed-forward networks (FFNs) that are disproportionately activated in such cases. Building on this insight, we propose Parametric Knowledge Muting through FFN Suppression (ParamMute), a framework that improves contextual faithfulness by suppressing the activation of unfaithfulness-associated FFNs and calibrating the model toward retrieved knowledge. To evaluate our approach, we introduce CoFaithfulQA, a benchmark specifically designed to evaluate faithfulness in scenarios where internal knowledge conflicts with accurate external evidence. Experimental results show that ParamMute significantly enhances faithfulness across both CoFaithfulQA and the established ConFiQA benchmark, achieving substantial reductions in reliance on parametric memory. These findings underscore the importance of mitigating internal knowledge dominance and provide a new direction for improving LLM trustworthiness in RAG.

Unlike MLPs, Kolmogorov-Arnold Networks (KANs) expose explicit learnable edge functions on every connection, enabling mechanistic explanation in time-series forecasting. This paper introduces Temporal Functional Circuits, a framework that transforms KAN edge functions from latent visualizations into faithful, temporally grounded explanations. Built on a gated residual KAN that decomposes forecasts into a linear base and a sparsely activated KAN correction, the framework (i) maps each edge to input lags via output-aware attribution, (ii) ranks edges by learned activation range, and (iii) validates faithfulness through edge-level interventions including zeroing and spline removal. Removing the learned B-spline component while retaining the base SiLU term degrades forecasts, providing evidence that the spline shape itself carries predictive value beyond the base activation. On four synthetic regimes of increasing complexity, the learned gate opens progressively wider as signal complexity grows. On regime-switching signals, gated KAN achieves 59% lower MSE than linear-only models. Across eight benchmarks, the gated architecture is competitive with linear, attention, and MLP alternatives, while providing interpretable edge functions that MLP-based corrections cannot offer.

Many of the causal discovery methods rely on the faithfulness assumption to guarantee asymptotic correctness. However, the assumption can be approximately violated in many ways, leading to sub-optimal solutions. Although there is a line of research in Bayesian network structure learning that focuses on weakening the assumption, such as exact search methods with well-defined score functions, they do not scale well to large graphs. In this work, we introduce several strategies to improve the scalability of exact score-based methods in the linear Gaussian setting. In particular, we develop a super-structure estimation method based on the support of inverse covariance matrix which requires assumptions that are strictly weaker than faithfulness, and apply it to restrict the search space of exact search. We also propose a local search strategy that performs exact search on the local clusters formed by each variable and its neighbors within two hops in the superstructure. Numerical experiments validate the efficacy of the proposed procedure, and demonstrate that it scales up to hundreds of nodes with a high accuracy.