论文检索

Social Aspects · Accountability, Transparency and Interpretability

Zheng Zhao, Yeskendir Koishekenov, Xianjun Yang, Naila Murray, Nicola Cancedda

Current Chain-of-Thought (CoT) verification methods predict reasoning correctness based on outputs (black-box) or activations (gray-box), but offer limited insight into \textit{why} a computation fails. We introduce a white-box method: \textbf{Circuit-based Reasoning Verification (CRV)}. We hypothesize that attribution graphs of correct CoT steps, viewed as \textit{execution traces} of the model's latent reasoning circuits, possess distinct structural fingerprints from those of incorrect steps. By training a classifier on structural features of these graphs, we show that these traces contain a powerful signal of reasoning errors. Our white-box approach yields novel scientific insights unattainable by other methods. (1) We demonstrate that structural signatures of error are highly predictive, establishing the viability of verifying reasoning directly via its computational graph. (2) We find these signatures to be highly domain-specific, revealing that failures in different reasoning tasks manifest as distinct computational patterns. (3) We provide evidence that these signatures are not merely correlational; by using our analysis to guide targeted interventions on individual transcoder features, we successfully correct the model's faulty reasoning. Our work shows that, by scrutinizing a model's computational process, we can move from simple error detection to a deeper, causal understanding of LLM reasoning.

Social Aspects · Accountability, Transparency and Interpretability

Junqi Jiang, Francesco Leofante, Antonio Rago, Francesca Toni

Counterfactual explanations (CEs) provide recourse recommendations for individuals affected by algorithmic decisions. A key challenge is generating CEs that are robust against various perturbation types (e.g. input and model perturbations) while simultaneously satisfying other desirable properties. These include plausibility, ensuring CEs reside on the data manifold, and diversity, providing multiple distinct recourse options for single inputs. Existing methods, however, mostly struggle to address these multifaceted requirements in a unified, model-agnostic manner. We address these limitations by proposing a novel generative framework. First, we introduce the Label-conditional Gaussian Mixture Variational Autoencoder (L-GMVAE), a model trained to learn a structured latent space where each class label is represented by a set of Gaussian components with diverse, prototypical centroids. Building on this, we present LAPACE (LAtent PAth Counterfactual Explanations), a model-agnostic algorithm that synthesises entire paths of CE points by interpolating from inputs' latent representations to those learned latent centroids. This approach inherently ensures robustness to input changes, as all paths for a given target class converge to the same fixed centroids. Furthermore, the generated paths provide a spectrum of recourse options, allowing users to navigate the trade-off between proximity and plausibility while also encouraging robustness against model changes. In addition, user-specified actionability constraints can also be easily incorporated via lightweight gradient optimisation through the L-GMVAE's decoder. Comprehensive experiments show that LAPACE is computationally efficient and achieves competitive performance across eight quantitative metrics.

Social Aspects · Accountability, Transparency and Interpretability

Matthias Wilms, Sascha Xu, Jilles Vreeken

Mixture models excel at decomposing complex, multi-modal distributions into simpler probabilistic components, but provide no insight into the conditions under which these components arise. We introduce explainable mixture models (XMM), a framework that pairs each mixture component with a human-interpretable rule over descriptive features. This enables mixtures that are not only statistically expressive but also transparently grounded in the underlying data. We formalize the problem and examine conditions under which an XMM exactly captures a target distribution. We then propose a scalable, differentiable learning procedure for discovering sets of rules. Experiments on synthetic and real-world datasets demonstrate that our method discovers interesting sub-populations in both univariate and multivariate settings, offering interpretable insights into the structure of complex distributions.

Social Aspects · Accountability, Transparency and Interpretability

Hanqi Yan, Hainiu Xu, Siya Qi, Shu Yang, Yulan He

With the growing accessibility and wide adoption of large language models, concerns about their safety and alignment with human values have become paramount. In this paper, we identify a concerning phenomenon: Reasoning-Induced Misalignment (RIM), in which misalignment emerges when reasoning capabilities strengthened—particularly when specific types of reasoning patterns are introduced during inference or training. Beyond reporting this vulnerability, we provide the first mechanistic account of its origins. Through representation analysis, we find that certain attention heads diverge from CoT tokens, modulating rationalization to enable refusal during generation. During training, we find significantly higher activation entanglement between reasoning and safety in safety-critical neurons than in control neurons, particularly after fine-tuning with those identified reasoning patterns. This entanglement strongly correlates with catastrophic forgetting, providing a neuron-level explanation for RIM.

Social Aspects · Accountability, Transparency and Interpretability

Shuangqi Li, Hieu Le, Jingyi Xu, Mathieu Salzmann

We study gradient-based data attribution, aiming to identify which training examples most influence a given output. Existing methods for this task either treat network parameters uniformly or rely on implicit weighting derived from Hessian approximations, which do not fully model functional heterogeneity of network parameters. To address this, we propose a method to explicitly learn parameter importance weights directly from data, without requiring annotated labels. Our approach improves attribution accuracy across diverse tasks, including image classification, language modeling, and diffusion, and enables fine-grained attribution for concepts like subject and style.

Social Aspects · Accountability, Transparency and Interpretability

Sayash Kapoor, Benedikt Stroebl, Peter Kirgis, Nitya Nadgir, Zachary Siegel, Boyi Wei, Tianci Xue, Ziru Chen, Felix Chen, Saiteja Utpala 等

AI agents have been developed for complex real-world tasks from coding to customer service. But AI agent evaluations suffer from many challenges that undermine our understanding of how well agents really work (Figure 1). We introduce the Holistic Agent Leaderboard (HAL) to address these challenges. We make three main contributions. First, we provide a standardized evaluation harness that orchestrates parallel evaluations across hundreds of VMs, reducing evaluation time from weeks to hours while eliminating common implementation bugs. Second, we conduct three-dimensional analysis spanning models, scaffolds, and benchmarks. We validate the harness by conducting 21,730 agent rollouts across 9 models and 9 benchmarks in coding, web navigation, science, and customer service with a total cost of about $40,000. Our analysis reveals surprising insights, such as higher reasoning effort reducing accuracy in the majority of runs. Third, we use LLM-aided log inspection to uncover previously unreported behaviors, such as searching for the benchmark on HuggingFace instead of solving a task, or misusing credit cards in flight booking tasks. We share all agent logs, comprising 2.5B tokens of language model calls, to incentivize further research into agent behavior. By standardizing how the field evaluates agents and addressing common pitfalls in agent evaluation, we hope to shift the focus from agents that ace benchmarks to agents that work reliably in the real world.

Social Aspects · Accountability, Transparency and Interpretability

Subhodip Panda, Dhruv Tarsadiya, Shashwat Sourav, Prathosh AP, Sai Karimireddy

Influence estimation methods promise to explain and debug machine learning by estimating the impact of individual samples on the final model. Yet, existing methods collapse under training randomness: the same example may appear critical in one run and irrelevant in the next. Such instability undermines their use in data curation or cleanup since it is unclear if we indeed deleted/kept the correct datapoints. To overcome this, we introduce *f-influence* -- a new influence estimation framework grounded in hypothesis testing that explicitly accounts for training randomness, and establish desirable properties that make it suitable for reliable influence estimation. We also design a highly efficient algorithm *f*-*IN*fluence *E*stimation (**f-INE**) that computes f-influence in a **in a single training run**. Finally, we scale up f-INE to estimate influence of instruction tuning data on Llama 3.1 8B and show it can reliably detect poisoned samples that steer model opinions, demonstrating its utility for data cleanup and attributing model behavior.

Social Aspects · Accountability, Transparency and Interpretability

Fabian Fumagalli, R. Teal Witter, Christopher Musco

Shapley values have emerged as a central game-theoretic tool in explainable AI (XAI). However, computing Shapley values exactly requires $2^d$ game evaluations for a model with $d$ features. Lundberg and Lee's KernelSHAP algorithm has emerged as a leading method for avoiding this exponential cost. KernelSHAP approximates Shapley values by approximating the game as a linear function, which is fit using a small number of game evaluations for random feature subsets. In this work, we extend KernelSHAP by approximating the game via higher degree polynomials, which capture non-linear interactions between features. Our resulting PolySHAP method yields empirically better Shapley value estimates for various benchmark datasets, and we prove that these estimates are consistent. Moreover, we connect our approach to paired sampling (antithetic sampling), a ubiquitous modification to KernelSHAP that improves empirical accuracy. We prove that paired sampling outputs exactly the same Shapley value approximations as second-order PolySHAP, without ever fitting a degree 2 polynomial. To the best of our knowledge, this finding provides the first strong theoretical justification for the excellent practical performance of the paired sampling heuristic.

Social Aspects · Accountability, Transparency and Interpretability

Seokhun Park, Choeun Kim, Jihu Lee, Yunseop Shin, Insung Kong, Yongdai Kim

With the increasing demand for interpretability in machine learning, functional ANOVA decomposition has gained renewed attention as a principled tool for breaking down high-dimensional function into low-dimensional components that reveal the contributions of different variable groups. Recently, Tensor Product Neural Network (TPNN) has been developed and applied as basis functions in the functional ANOVA model, referred to as ANOVA-TPNN. A disadvantage of ANOVA-TPNN, however, is that the components to be estimated must be specified in advance, which makes it difficult to incorporate higher-order TPNNs into the functional ANOVA model due to computational and memory constraints. In this work, we propose Bayesian-TPNN, a Bayesian inference procedure for the functional ANOVA model with TPNN basis functions, enabling the detection of higher-order components with reduced computational cost compared to ANOVA-TPNN. We develop an efficient MCMC algorithm and demonstrate that Bayesian-TPNN performs well by analyzing multiple benchmark datasets. Theoretically, we prove that the posterior of Bayesian-TPNN is consistent.

Social Aspects · Accountability, Transparency and Interpretability

Yein Park, Jungwoo Park, Jaewoo Kang

Large language models (LLMs), despite being safety-aligned, exhibit brittle refusal behaviors that can be circumvented by simple linguistic changes. As tense jailbreaking demonstrates that models refusing harmful requests often comply when rephrased in past tense, a critical generalization gap is revealed in current alignment methods whose underlying mechanisms are poorly understood. In this work, we introduce Activation-Scaling Guard (ASGuard), an insightful, mechanistically-informed framework that surgically mitigates this specific vulnerability. In the first step, we use circuit analysis to identify the specific attention heads causally linked to the targeted jailbreaking such as a tense-changing attack. Second, we train a precise, channel-wise scaling vector to recalibrate the activation of tense vulnerable heads. Lastly, we apply it into a "preventative fine-tuning", forcing the model to learn a more robust refusal mechanism. Across four LLMs, ASGuard effectively reduces the attack success rate of targeted jailbreaking while preserving general capabilities and minimizing over refusal, achieving a Pareto-optimal balance between safety and utility. Our findings underscore how adversarial suffixes suppress the propagation of the refusal-mediating direction, based on mechanistic analysis. Furthermore, our work showcases how a deep understanding of model internals can be leveraged to develop practical, efficient, and targeted methods for adjusting model behavior, charting a course for more reliable and interpretable AI safety.

Social Aspects · Accountability, Transparency and Interpretability

An Vo, Khai-Nguyen Nguyen, Mohammad Reza Taesiri, THI TUONG VY DANG, Anh Nguyen, Daeyoung Kim

Large language models (LLMs) memorize a vast amount of prior knowledge from the Internet that helps them on downstream tasks but also may notoriously sway their outputs toward wrong or biased answers. In this work, we test how the knowledge of popular subjects hurts the accuracy of vision language models (VLMs) on standard, objective visual tasks of counting and identification. We find that state-of-the-art VLMs are strongly biased (e.g., unable to recognize that a 4th stripe has been added to a 3-stripe Adidas logo), scoring an average of 17.05% accuracy in counting (e.g., counting stripes in an Adidas-like logo) across 7 diverse domains spanning animals, logos, chess, game boards, optical illusions, and patterned grids. Removing image backgrounds nearly doubles accuracy (by 21.09 points), revealing that background visual cues trigger these biased responses. Further analysis of VLMs' reasoning patterns shows that counting accuracy initially rises with thinking tokens, reaching ~40%, before declining with model overthinking. Our work presents an interesting failure mode in VLMs and a human-supervised automated framework for testing VLM biases. Code and data are available at: vlmsarebiased.github.io.

Social Aspects · Accountability, Transparency and Interpretability

Yuhui Wang, Changjiang Li, Guangke Chen, Jiacheng Liang, Ting Wang

Large reasoning models (LRMs) exhibit unprecedented capabilities in solving complex problems through Chain-of-Thought (CoT) reasoning. However, recent studies reveal that their final answers often contradict their own reasoning traces. We hypothesize that this inconsistency stems from two competing mechanisms for generating answers: CoT reasoning and memory retrieval. To test this hypothesis, we conduct controlled experiments that challenge LRMs with misleading cues during reasoning and/or corrupted answers during retrieval. Our results across models and datasets confirm that both mechanisms operate simultaneously, with their relative dominance influenced by multiple factors: problem domains, model scales, and fine-tuning approaches (e.g., reinforcement learning vs. distillation). The findings reveal a critical limitation in current reasoning fine-tuning paradigms: models can exploit the retrieval mechanism as a shortcut, effectively "hacking" the reward signal and undermining genuine reasoning development. To address this challenge, we introduce FARL, a novel fine-tuning framework that integrates memory unlearning with reinforcement learning. By carefully suppressing retrieval shortcuts during the fine-tuning process, FARL promotes reasoning-dominant behavior and enhances generalizable reasoning capabilities. The code is available at https://github.com/ZJUWYH/FARL.

Social Aspects · Accountability, Transparency and Interpretability

Ruimeng Ye, Zihan Wang, Zinan Ling, Yang Xiao, Manling Li, Xiaolong Ma, Bo Hui

Humans shift between different personas depending on social context. Large Language Models (LLMs) demonstrate a similar flexibility in adopting different personas and behaviors. Existing approaches, however, typically adapt such behavior through external knowledge such as prompting, retrieval-augmented generation (RAG), or fine-tuning. We ask: do LLMs really need external context or parameters to adapt to different behaviors, or do they already have such knowledge embedded in their parameters? In this work, we show that LLMs already contain persona-specialized subnetworks in their parameter space. Using small calibration datasets, we identify distinct activation signatures associated with different personas. Guided by these statistics, we develop a masking strategy that isolates lightweight persona subnetworks. Building on the findings, we further discuss: how can we discover opposing subnetworks from the model that lead to binary-opposing personas, such as introvert-extrovert? To further enhance separation in binary opposition scenarios, we introduce a contrastive pruning strategy that identifies parameters responsible for the statistical divergence between opposing personas. Our method is entirely training-free and relies solely on the language model's existing parameter space. Across diverse evaluation settings, the resulting subnetworks exhibit significantly stronger persona alignment than baselines that require external knowledge while being more efficient. Our findings suggest that diverse human-like behaviors are not merely induced in LLMs, but are already embedded in their parameter space—pointing toward a new perspective on controllable and interpretable personalization in large language models. Our code is available at https://github.com/Ruimeng-Ye/Persona.git.

Social Aspects · Accountability, Transparency and Interpretability

Qi Zhang, Yifei Wang, Xiaohan Wang, Jiajun Chai, Guojun Yin, Wei Lin, Yisen Wang

In recent years, pre-trained large language models have achieved remarkable success across diverse tasks. Besides the pivotal role of self-supervised pre-training, their effectiveness in downstream applications also depends critically on the post-training process, which adapts models to task-specific data and objectives. However, this process inevitably introduces model shifts that can influence performance in different domains, and how such shifts transfer remains poorly understood. To open up the black box, we propose the SAE-based Transferability Score (STS), a new metric that leverages sparse autoencoders (SAEs) to forecast post-training transferability. Taking supervised fine-tuning as an example, STS identifies shifted dimensions in SAE representations and calculates their correlations with downstream domains, enabling reliable estimation of transferability \textit{before} fine-tuning. Extensive experiments across multiple models and domains show that STS accurately predicts the transferability of supervised fine-tuning, achieving Pearson correlation coefficients above 0.7 with actual performance changes. Beyond this, we take an initial step toward extending STS to reinforcement learning. We believe that STS can serve as an {\color{black} interpretable} tool for guiding post-training strategies in LLMs. Code is available at \url{https://github.com/PKU-ML/STS}.

Social Aspects · Accountability, Transparency and Interpretability

Taisei Yamamoto, Ryoma Kumon, Danushka Bollegala, Hitomi Yanaka

As large language models (LLMs) are increasingly deployed worldwide, ensuring their fair and comprehensive cultural understanding is important. However, LLMs exhibit cultural bias and limited awareness of underrepresented cultures, while the mechanisms underlying their cultural understanding remain underexplored. To fill this gap, we conduct a neuron-level analysis to identify neurons that drive cultural behavior, introducing a gradient-based scoring method with additional filtering for precise refinement. We identify culture-general neurons contributing to cultural understanding regardless of cultures, and culture-specific neurons tied to an individual culture. Culture-general and culture-specific neurons account for less than 1% of all neurons and are concentrated in shallow to middle MLP layers. We validate their role by showing that suppressing them substantially degrades performance on cultural benchmarks (by up to 30%), while performance on general natural language understanding (NLU) benchmarks remains largely unaffected. Moreover, we show that culture-specific neurons support knowledge of not only the target culture, but also related cultures. Finally, we demonstrate that training on NLU benchmarks can diminish models' cultural understanding when we update modules containing many culture-general neurons. These findings provide insights into the internal mechanisms of LLMs and offer practical guidance for model training and engineering. Our code is available at https://github.com/ynklab/CULNIG

Social Aspects · Accountability, Transparency and Interpretability

Yuntai Bao, Xuhong Zhang, Jintao Chen, Ge Su, Yuxiang Cai, Hao Peng, SUN Bing, Haiqin Weng, Liu Yan, Jianwei Yin

Intervention-based model steering offers a lightweight and interpretable alternative to prompting and fine-tuning. However, by adapting strong optimization objectives from fine-tuning, current methods are susceptible to overfitting and often underperform, sometimes generating unnatural outputs. We hypothesize that this is because effective steering requires the faithful identification of internal model mechanisms, not the enforcement of external preferences. To this end, we build on the principles of *distributed alignment search (DAS)*, the standard for causal variable localization, to propose a new steering method: **Concept DAS (CDAS)**. While we adopt the core mechanism of DAS, *distributed interchange intervention (DII)*, we introduce a novel distribution matching objective tailored for the steering task by aligning intervened output distributions with counterfactual distributions. CDAS differs from prior work in two main ways: first, it learns interventions via weak-supervised distribution matching rather than probability maximization; second, it uses DIIs that naturally enable bi-directional steering and allow steering factors to be derived from data, reducing the effort required for hyperparameter tuning and resulting in more faithful and stable control. On AxBench, a large-scale model steering benchmark, we show that CDAS does not always outperform preference-optimization methods but may benefit more from increased model scale. In two safety-related case studies, overriding refusal behaviors of safety-aligned models and neutralizing a chain-of-thought backdoor, CDAS achieves systematic steering while maintaining general model utility. These results indicate that CDAS is complementary to preference-optimization approaches and conditionally constitutes a robust approach to intervention-based model steering.

Social Aspects · Accountability, Transparency and Interpretability

Gabriel Laberge, Ola Ahmad

Post-hoc explanations aim at understanding which input features (or groups thereof) are the most impactful toward certain model decisions. Many such methods have been proposed (ArchAttribute, Occlusion, SHAP, RISE, LIME, Integrated Gradient) and it is hard for practitioners to understand the differences between them. Even worse, faithfulness metrics, often used to quantitatively compare explanation methods, also exhibit inconsistencies. To address these issues, recent work has unified explanation methods through the lens of Functional Decomposition. We extend such work to scenarios where input features are partitioned into groups (e.g. pixel patches) and prove that disagreements between explanation methods and faithfulness metrics are caused by between-group interactions. Crucially, getting rid of between-group interactions leads to a single explanation that is optimal according to all faithfulness metrics. We finally show how to reduce the disagreements by grouping features on tabular/image data.

Social Aspects · Accountability, Transparency and Interpretability

Borui Yang, Mingxuan Ma, Liyao Xiang, Nan Chen, Xin Zhang, Linghe Kong, Xinghao Jiang

Code language models (LMs) represent valuable intellectual property (IP) as their training involves immense investments, including large-scale code corpora, proprietary annotations, extensive computational resources, and specialized designs. Hence the threat of model IP infringements such as unauthorized redistribution or model theft has become increasingly concerning. While neural network watermarking has been widely studied as a measure to support model ownership verification, watermarking code LMs is particularly challenging due to the seemingly conflicting requirements of code generation: adhering to strict syntactic rules and semantic consistency while allowing flexible changes to embed watermarks, keeping high fidelity of the generated content while being robust to extraction attacks, etc. To resolve the issues, we propose CodeGenGuard, a watermarking framework for code LMs. CodeGenGuard leverages semantic-preserving transformations (SPTs) to encode the watermark and incorporates a dead-code-based data augmentation pipeline to diversify SPT patterns. To improve robustness, we incorporate an efficient dual-LoRA shadow training scheme and an optimizable trigger prompt that learns to extract watermark from both the watermarked and the shadow models. As most SPTs take place in specific contexts, we implant auxiliary prompts during verification to encourage the generation of the context, further enhancing the detection rate. Evaluation results on representative code generation models demonstrate that CodeGenGuard achieves superior watermarking performance to the state-of-the-art.

Social Aspects · Accountability, Transparency and Interpretability

Sara Kangaslahti, Elan Rosenfeld, Naomi Saphra

Loss curves are smooth during most of model training, so visible discontinuities stand out as possible conceptual breakthroughs. Studying these breakthroughs enables a deeper understanding of learning dynamics, but only when they are properly identified. This paper argues that similar breakthroughs occur frequently throughout training but they are obscured by a loss metric that collapses all variation into a single scalar. To find these hidden transitions, we introduce POLCA, a method for decomposing changes in loss along arbitrary bases of the low-rank training subspace. We use our method to identify clusters of samples that share similar changes in loss during training, disaggregating the overall loss into that of smaller groups of conceptually similar data. We validate our method on synthetic arithmetic and natural language tasks, showing that POLCA recovers clusters that represent interpretable breakthroughs in the model's capabilities. We demonstrate the promise of these hidden phase transitions as a tool for unsupervised interpretability.

Social Aspects · Accountability, Transparency and Interpretability

Uzay Macar, Paul Bogdan, Senthooran Rajamanoharan, Neel Nanda

We argue that interpreting reasoning models from a single chain-of-thought (CoT) is fundamentally inadequate. To understand computation and causal influence, one must study reasoning as a distribution of possible trajectories elicited by a given prompt. We approximate this distribution via on-policy resampling and use it to answer concrete questions about the causes of model decisions. First, when a model states a reason for its action, does that reason actually cause the action? In agentic misalignment scenarios where models seemingly blackmail to preserve themselves, we resample specific sentences to measure their downstream effects. We find that normative self-preservation sentences have unusually small and non-resilient causal impact on the final decision across models, indicating they are not a meaningful driver of blackmail. Second, are handwritten edits to CoT sufficient for steering reasoning? We find that off-policy sentence insertions common in earlier literature yield small and unstable effects in decision-making tasks, whereas on-policy resampling produces larger and more consistent effects. Third, how do we attribute causal influence when models modify their plans or correct prior errors during reasoning? We introduce a resilience metric and counterfactual importance that repeatedly resample to remove sentences such that similar content doesn't reappear downstream. Critical planning statements resist removal but have large effects when successfully eliminated. Fourth, what can our methods, which focus on the mechanistic roles of CoT, teach us about unfaithful reasoning? Adapting causal mediation analysis, we edit hint pathways mid-trajectory and find that prompt hints exert smooth and cumulative influences rather than single-step pivots. Hidden information can influence the trajectory of reasoning by shifting what decisions are made at different junctures in a CoT, and these biases can be modeled and quantified with resampling. Overall, studying distributions via resampling enables reliable causal analysis, clearer narratives of model reasoning, and principled guidance for CoT interventions.