Vision-Language Models (VLMs) have made significant progress in static perception, but their ability to understand dynamic task-oriented reasoning remains unclear. Existing benchmarks mainly focus on static spatial relationships and lack systematic assessment of dynamic reasoning capabilities. To this end, we propose SpatialLogic-Bench, a novel benchmark designed to evaluate VLMs’ understanding of spatiotemporal logic and their ability to assess task progress. The benchmark assesses two critical capabilities: first, fine-grained visual discrimination to accurately perceive subtle physical changes between state frames; second, the logical capacity to connect these changes to task goals and judge whether they indicate progress. To mitigate temporal dependency biases, we introduce a dual-task paradigm, presenting image pairs in both chronological and reversed orders while keeping task descriptions consistent. We construct a multi-scale evaluation system by varying time intervals between frames: smaller intervals test the model's fine-grained perception, while larger intervals demand more sophisticated logical inference. Empirical evaluation reveals that most VLMs experience significant performance degradation on tasks presented in inverse chronological order, indicating an over-reliance on temporal cues rather than robust reasoning abilities. SpatialLogic-Bench clearly exposes critical limitations in current models and provides valuable guidance for improving dynamic spatial perception capabilities.
论文检索
输入标题、作者或关键词，从 100,903 篇学术成果中精准定位
Socrates or Smartypants: Testing Logic Reasoning Capabilities of Large Language Models with Logic Programming-Based Test Oracles
PDF ↗Large Language Models (LLMs) have achieved significant progress in language understanding and reasoning. Evaluating and analyzing their logical reasoning abilities has therefore become essential. However, existing datasets and benchmarks are often limited to overly simplistic, unnatural, or contextually constrained examples. In response to the growing demand, we introduce SMARTYPAT-BENCH, a challenging, naturally expressed, and systematically labeled benchmark derived from real-world high-quality Reddit posts containing subtle logical fallacies. Unlike existing datasets and benchmarks, it provides more detailed annotations of logical fallacies and features more diverse data. To further scale up the study and address the limitations of manual data collection and labeling, such as fallacy-type imbalance and labor-intensive annotation, we introduce SMARTYPAT, an automated framework powered by logic programming-based oracles. SMARTYPAT utilizes Prolog rules to systematically generate logically fallacious statements, which are then refined into fluent natural language sentences by LLMs, ensuring precise fallacy rep- resentation. Extensive evaluation demonstrates that SMARTYPAT produces fallacies comparable in subtlety and quality to human-generated content and significantly outperforms baseline methods. Finally, experiments reveal insights into LLM capabilities, highlighting that while excessive reasoning steps hinder fallacy detection accuracy, structured reasoning enhances fallacy categorization performance.
Geometry Problem Solving has become a hot topic these years due to its complexity of enabling the machine with geometric abstraction, multi-modal reasoning and mathematical capabilities. Majority of research works place their attention on the fusion of multi-modal data or the synergistic combination of neural and symbolic systems for performance improvement. However, their neglect of the unique characteristics of geometric diagrams, which distinguish them from natural images, impedes the further exploring of critical information in geometric diagrams. In this work, we introduce the novel concept of geo-graph and propose the Geo-Graph Geometry Problem Solving model which encodes the geometric diagram from a new perspective. The geo-graph is designed to include semantic, structural and spatial information in the diagram, which is crucial to subsequent problem reasoning stage. To facilitate the model's comprehension of the actual layout of geometric diagram, spatial and connecting attentions are devised to serve as intrinsic knowledge guidance for feature propagation. An extra cross-modal attention is used as external guidance to instruct the encoding of geo-graph to be related to specific problem target. Fused multi-modal features are then sent into a commonly used encoder-decoder framework for final solution generation. The model is first trained with three carefully designed pre-training tasks to establish its fundamental knowledge of geo-graph, leveraging numerous varied samples generated through a geo-graph-based augmentation method. Experiments on popular geometry problem solving datasets demonstrate the effectiveness and superiority of our model for geometric diagram encoding.
Establishing convergent semantics for weighted argumentation graphs is a long-standing fundamental issue. Particularly, it is challenging to develop convergent semantics for weighted bipolar argumentation graphs (wBAG), which include both support and attack relations on weighted arguments. Existing semantics in the literature are not general enough in the sense that they only apply to acyclic graphs or special cyclic cases. In this paper, we provide an elegant solution to this issue by adopting the so-called bilateral gradual semantics, so that the strength of arguments can be defined as the limits of iterative functions that always converge for any wBAG including cyclic ones. A preliminary experimental analysis shows that our semantics appear quite efficient in calculating argument strength. Overall, this paper offers a solid and promising foundation for weighted bipolar argumentation in theoretical and practical aspects.
Despite their widespread use in quantum computing and physics, the relative strengths and weaknesses of Matrix Product States (MPS), Decision Diagrams (DDs), and Restricted Boltzmann Machines (RBMs) remains poorly understood. We analytically compare the succinctness of these quantum state representations and analyze the complexity of key operations on them. To overcome shortcomings of the tractability measure, we introduce `rapidity' conditions that identify when non-canonical representations efficiently simulate each other. Our results reveal that: 1. Most DD variants are redundant with respect to MPS in a strong sense; MPS is more rapid. 2. Only one DD variant, called LIMDD, and RBM have succinctness incomparable to MPS. 3. LIMDD and RBM seem to achieve this by sacrificing tractability of counting queries, as shown by a metatheorem on the conditional hardness of these queries.
Machine unlearning has emerged as a promising approach to remove specific knowledge from large language models (LLMs), especially for safety-critical applications. However, existing representation-based methods lack guidance for selecting representation locations to unlearn (RMU), thus lacking precision in unlearning, while probability-based methods are vulnerable to fine-tuning attacks which use unrelated and safe data to fine-tune models. To address these problems, this paper presents an adaptive knowledge guidance and memory perturbation mechanisms, called ALMPU (Adaptive Localized Memory Perturbation Unlearning) which addresses the lack of knowledge guidance in representation-based unlearning methods and mitigates the impact of fine-tuning attacks on unlearned models. Specifically, we apply scaling factors to attention heads and select the most sensitive ones as knowledge guidance. Guided by the previous knowledge localization, we integrate enhanced memory perturbation—which forces the model to preserve specific knowledge—into the standard representation-based unlearning process at these sensitive positions. Through this perturbation mechanism, the model achieves more thorough elimination of the target knowledge. By adding interventions to selected attention heads and explicitly optimizing against fine-tuning attacks during the unlearning process, ALMPU creates a controlled divergence from the original model that is inherently resistant to relearning attempts. Experimental evaluation on the WMDP benchmark demonstrates that ALMPU consistently outperforms baseline methods across different scales of fine-tuning attacks.
Delphi: A Neuro-Symbolic Framework for Individualized, Safe and Interpretable Treatment Recommendation
PDF ↗Clinical reinforcement learning (RL) holds promise for treatment recommendation but remains hindered by black-box decision processes, limited safety guarantees, and lack of individualized reasoning. We introduce Delphi Engine, the first fully trainable neuro-symbolic causal RL framework for dynamic treatment planning, designed to answer three core clinical questions in real time: Why this action? Why is it safe? Why for this patient? Specifically, Delphi integrates: (1) causality-aware state modeling using discretized physiological variables and subtype-specific causal graphs; (2) adaptive symbolic rule constraints, combining clinical guidelines and behavior-derived rules into soft differentiable logic; and (3) interpretable decision fusion, where actions are selected based on joint neural-symbolic Q-values and explained via structured LLM-based justifications. We evaluate Delphi on the MIMIC-III sepsis cohort using both standard off-policy evaluations (WIS↑1.47, DR↑1.29, RMSE↓0.207) and the first blinded physician evaluation of an explainable RL system in healthcare. Delphi consistently outperforms historical physicians' treatments in safety (+10.4%), understandability (+8.9%), and adoption rate (+5.75%) across six clinical axes. These results highlight Delphi’s potential as a safe, interpretable, and patient-specific AI assistant for critical care medicine.
HierSearch: A Hierarchical Enterprise Deep Search Framework Integrating Local and Web Searches
PDF ↗Recently, large reasoning models have demonstrated strong mathematical and coding abilities, and deep search leverages their reasoning capabilities in challenging information retrieval tasks. Existing deep search works are generally limited to a single knowledge source, either local or the Web. However, enterprises often require private deep search systems that can leverage search tools over both local and the Web corpus. Simply training an agent equipped with multiple search tools using flat reinforcement learning (RL) is a straightforward idea, but it has problems such as low training data efficiency and poor mastery of complex tools. To address the above issue, we propose a hierarchical agentic deep search framework, HierSearch, trained with hierarchical RL. At the low level, a local deep search agent and a Web deep search agent are trained to retrieve evidence from their corresponding domains. At the high level, a planner agent coordinates low-level agents and provides the final answer. Moreover, to prevent direct answer copying and error propagation, we design a knowledge refiner that filters out hallucinations and irrelevant evidence returned by low-level agents. Experiments show that HierSearch achieves better performance compared to flat RL, and outperforms various deep search and multi-source retrieval-augmented generation baselines in six benchmarks across general, finance, and medical domains.
Description Logics with Two Types of Definite Descriptions: Complexity, Expressiveness, and Automated Deduction
PDF ↗Definite descriptions are expressions of the form „the unique x satisfying property C,” which allow reference to objects through their distinguishing characteristics. They play a crucial role in ontology and query languages, offering an alternative to proper names (IDs), which lack semantic content and serve merely as placeholders. In this paper, we introduce two extensions of the well-known description logic ALC with local and global definite descriptions, denoted ALCiL and ALCiG, respectively. We define appropriate bisimulation notions for these logics, enabling an analysis of their expressiveness. We show that although both logics share the same tight ExpTime complexity bounds for concept and ontology satisfiability, ALCiG is strictly more expressive than ALCiL. Moreover, we present tableau-based decision procedures for satisfiability in both logics, provide their implementation, and report on a series of experiments. The empirical results demonstrate the practical utility of the implementation and reveal interesting correlations between performance and structural properties of the input formulas.
Knowledge editing aims to update specific knowledge in Large Language Models (LLMs) without retraining the entire model. However, existing methods generally struggle to manage the ripple effects of knowledge updates, particularly in multi-hop reasoning tasks, where conflicts between old and new information often lead to shifts in reasoning chains and degraded consistency. To address this issue, a ripple-aware knowledge editing framework, namely EchoEdit, is proposed. EchoEdit introduces the RippleGraph to explicitly model potentially affected knowledge regions and employs a RippleRule generator to dynamically produce diffusion rules, precisely constraining knowledge propagation. Furthermore, we distill a Chain-of-Thought (CoT) planner from an external teacher model, which decouples complex reasoning chain planning into RippleGraph-guided reasoning, thereby alleviating the reasoning burden on low-resource LLMs in multi-hop tasks. Experimental results on the MQuAKE and RIPPLEEDITS multi-hop reasoning benchmarks demonstrate that EchoEdit significantly outperforms existing mainstream methods, effectively enhancing post-edit reasoning consistency and generalization capabilities.
Truth-tracking in collective reasoning systems is a core challenge in domains such as e-democracy, online deliberation, and citizen opinion polling. Our prior work introduced Opinion-Based Argumentation (OBA), a framework modeling both voting and argumentation, along with collective opinion semantics (COS) designed to select sets of arguments that are mutually coherent and aligned with agents' votes. In this paper, we first formally define the truth-tracking problem within OBA. We then introduce VAST, a comprehensive evaluation framework to systematically assess the epistemic adequacy of COS. Our empirical analysis, conducted using VAST, demonstrates substantial variation in their truth-tracking performance across diverse deliberative conditions.
Spatial understanding is a critical capability for LVLMs (Large Vision-Language Models) to advance embodied AI applications. Existing works primarily focus on enhancing spatial understanding within a single frame, i.e., injecting 3D spatial concepts into LVLMs under single coordinate system. However, such improvements struggle in real-world tasks that require consistent cross-view spatial reasoning. In this paper, we propose CVVG-Reasoner(Cross-View Visual Geometries) that lifts single-frame spatial comprehension to unified cross-view spatial understanding by mimicking human-like cross-view reasoning mechanisms. First, we introduce MV3DSR(Multi-View 3D Spatial Reasoning), a scalable pipeline for cross-view spatial reasoning data generation, and construct MV3DSR-Dataset, a large-scale dataset with diverse 3D cross-view reasoning tasks. Based on MV3DSR, we propose MV3DSR-Bench, a comprehensive benchmark for evaluating cross-view spatial reasoning capabilities. Second, we design a three-stage training strategy: the first two stages progressively equip the model with (1) fundamental spatial knowledge and (2) human-like cross-view reasoning patterns, while the final stage employs reinforcement learning to further boost its performance. Extensive experiments demonstrate that our CVVG-Reasoner significantly outperforms existing 3D LLMs(Large Language Models) and advanced LVLMs in cross-view tasks while maintaining robust performance on out-of-domain data. Ablations further reveal that injecting human-like reasoning patterns yields 44% performance gain, validating the effectiveness of our design.
Inductive Logic Programming (ILP) approaches like Meta-Interpretive Learning (MIL) can learn, from few examples, recursive logic programs with invented predicates that generalise well to unseen instances. This ability relies on a background theory and negative examples, both carefully selected with expert knowledge of a learning problem and its solutions. But what if such a problem-specific background theory or negative examples are not available? We formalise this question as a new setting for Self-Supervised ILP and present a new MIL algorithm that learns in the new setting from some positive labelled, and zero or more unlabelled examples, and automatically generates, and labels, new positive and negative examples during learning. We implement this algorithm in Prolog in a new MIL system, called Poker. We compare Poker to state-of-the-art MIL system Louise on experiments learning grammars for Context-Free and L-System languages from labelled, positive example strings, no negative examples, and just the terminal vocabulary of a language, seen in examples, as a first-order background theory. We introduce a new approach for the principled selection of a second-order background theory as a Second Order Definite Normal Form (SONF), sufficiently general to learn all programs in a class, thus removing the need for a backgound theory tailored to a learning task. We find that Poker's performance improves with increasing numbers of automatically generated examples while Louise, bereft of negative examples, over-generalises.
Time-frequency analysis (TFA) and mode decomposition for non-stationary signals are research hotspots in the field of signal processing. Current optimization-based decomposition methods require a good initial IF estimate. However, due to the Heisenberg uncertainty, achieving accurate ridge extraction from the time-frequency representation (TFR) necessitates empirical parameter adjustments. In this paper, we propose the TFD-Net framework, which takes time-series signals as inputs and adaptively conducts TFR construction and mode decomposition. Specifically, the framework integrates a physically interpretable TFA encoder and a query-based mode decomposition decoder. The highlights of this study include exploring the mathematical equivalence between deep convolutional operators and classical TFA methods. This enables the extraction of multi-scale features for TFR construction and mode separation in a data-driven manner, eliminating the need for signal-specific manual tuning and enhancing adaptability. Finally, simulated and real-world experiments demonstrate TFD-Net's superior performance over several state-of-the-art methods in complex signal processing.
We consider the problem of modifying a description logic concept in light of models represented as pointed interpretations. We call this setting model change, and distinguish three main kinds of changes: eviction, which consists of only removing models; reception, which incorporates models; and revision, which combines removal with incorporation of models in a single operation. We introduce a formal notion of revision and argue that it does not reduce to a simple combination of eviction and reception, contrary to intuition. We provide positive and negative results on the compatibility of eviction and reception for EL-bottom and ALC description logic concepts and on the compatibility of revision for ALC concepts.
One of the most important queries in knowledge compilation is weighted model counting (WMC), which has been applied to probabilistic inference on various models, such as Bayesian networks. In practical situations on inference tasks, the model's parameters have uncertainty because they are often learned from data, and thus we want to compute the degree of uncertainty in the inference outcome. One possible approach is to regard the inference outcome as a random variable by introducing distributions for the parameters and evaluate the variance of the outcome. Unfortunately, the tractability of computing such a variance is hardly known. Motivated by this, we consider the problem of computing the variance of WMC and investigate this problem's tractability. First, we derive a polynomial time algorithm to evaluate the WMC variance when the input is given as a structured d-DNNF. Second, we prove the hardness of this problem for structured DNNFs, d-DNNFs, and FBDDs, which is intriguing because the latter two allow polynomial time WMC algorithms. Finally, we show an application that measures the uncertainty in the inference of Bayesian networks. We empirically show that our algorithm can evaluate the variance of the marginal probability on real-world Bayesian networks and analyze the impact of the variances of parameters on the variance of the marginal.
In systems such as group calendars or collaborative platforms, agents make group commitments to future actions that must adapt as new facts or constraints emerge. We develop a formal framework for revising such group intentions in systems where coalitions adopt shared, temporally extended intentions represented in a logic based on Alternating-Time Temporal Logic with strategy contexts. After formulating coherence criteria for systems of group intentions, we establish representation theorems in the style of Katsuno and Mendelzon, showing that revision operators satisfy rationality postulates precisely when they can be represented by preorders on strategy profiles. These results extend classical revision theory by covering non-total preorders and a logic of higher expressive power. Altogether, the framework lays the groundwork for principled revision of group intentions in systems where both coordination and change are essential.
Structural measures of graphs, such as treewidth, are central tools in computational complexity resulting in efficient algorithms when exploiting the parameter. It is even known that modern SAT solvers work efficiently on instances of small treewidth. Since these solvers are widely applied, research interests in compact encodings into (Q)SAT for solving and to understand encoding limitations. Even more general is the graph parameter clique-width, which unlike treewidth can be small for dense graphs. Although algorithms are available for clique-width, little is known about encodings. We initiate the quest to understand encoding capabilities with clique-width by considering abstract argumentation, which is a robust framework for reasoning with conflicting arguments. It is based on directed graphs and asks for computationally challenging properties, making it a natural candidate to study computational properties. We design novel reductions from argumentation problems to (Q)SAT. Our reductions linearly preserve the clique-width, resulting in directed decomposition-guided (DDG) reductions. We establish novel results for all argumentation semantics, including counting. Notably, the overhead caused by our DDG reductions cannot be significantly improved under reasonable assumptions.
Knowledge Tracing (KT) aims to dynamically model a student’s mastery of knowledge concepts based on their historical learning interactions. Most current methods rely on single-point estimates, which cannot distinguish true ability from outburst or carelessness, creating ambiguity in judging mastery. To address this issue, we propose a Knowledge Mastery-State Disambiguation for Knowledge Tracing model (KeenKT), which represents a student’s knowledge state at each interaction using a Normal-Inverse-Gaussian (NIG) distribution, thereby capturing the fluctuations in student learning behaviors. Furthermore, we design an NIG-distance-based attention mechanism to model the dynamic evolution of the knowledge state. In addition, we introduce a diffusion-based denoising reconstruction loss and a distributional contrastive learning loss to enhance the model’s robustness. Extensive experiments on six public datasets demonstrate that KeenKT outperforms state-of-the-art KT models in terms of prediction accuracy and sensitivity to behavioral fluctuations. The proposed method yields the maximum AUC improvement of 5.85% and the maximum ACC improvement of 6.89%.
MyGram: Modality-aware Graph Transformer with Global Distribution for Multi-modal Entity Alignment
PDF ↗Multi-modal entity alignment aims to identify equivalent entities between two multi-modal Knowledge graphs by integrating multi-modal data, such as images and text, to enrich the semantic representations of entities. However, existing methods may overlook the structural contextual information within each modality, making them vulnerable to interference from shallow features. To address these challenges, we propose MyGram, a \textbf{m}odalit\textbf{y}-aware \textbf{gra}ph transformer with global distribution for \textbf{m}ulti-modal entity alignment. Specifically, we develop a modality diffusion learning module to capture deep structural contextual information within modalities and enable fine-grained multi-modal fusion. In addition, we introduce a Gram Loss that acts as a regularization constraint by minimizing the volume of a 4-dimensional parallelotope formed by multi-modal features, thereby achieving global distribution consistency across modalities. We conduct experiments on five public datasets. Results show that MyGram outperforms baseline models, achieving a maximum improvement of 4.8\% in Hits@1 on FBDB15K, 9.9\% on FBYG15K, and 4.3\% on DBP15K.