Artificial Intelligence · Cross-listed from cs.LG
Calendar-Structured Sparse Principal Component Analysis for Interpretable Multi-Periodic Electricity Consumption Profiles
Carlos Quesada-Granja, Tony Castillo-Calzadilla and Carlos Rizo-Maestre
Long-term electricity-consumption profiles exhibit several simultaneous periodic structures, including daily, weekly, and annual cycles. This work introduces Calendar-Structured Sparse Principal Component Analysis (Calendar-SPCA), a structured representation-learning method that incorporates this known multi-periodic geometry directly into a low-dimensional factorization. The method represents the feature domain as the Cartesian product of cyclic calendar axes and combines an L1 loading penalty with graph total variation, producing sparse, locally coherent, and directly interpretable latent factors. In this study, Calendar-SPCA is applied to the interpretable analysis of long-term electricity-consumption profiles and evaluated on two independent public smart-meter
datasets, GoiEner and Low Carbon London, with different population sizes and temporal resolutions. A factorial experiment characterizes the effects of sparsity and calendar coherence and examines robustness across sample size, latent dimensionality, and repeated fits. At rank 15, Calendar-SPCA retains 96.92% and 82.90% of the explained variance of rank-matched principal component analysis (PCA) in GoiEner and Low Carbon London, respectively, with mean loading sparsities of 61.95% and 81.50%. Comparisons with classical sparse PCA and sparse PCA with total variation (SPCA-TV) show that Calendar-SPCA organizes latent factors into interpretable structures over the daily, weekly, and annual calendar axes while preserving substantial low-rank information.
Read the full abstract ▾
Artificial Intelligence
OmniStepBench: Deployment-Aware Agent Evaluation across Modalities and Task Complexity
Yu Liu, Zhilin Liu, Zhiwei Yang et al. (+7 more)
As large
language models evolve from question-answering systems into general-purpose
agents, evaluation must move beyond static answer correctness to assess
multimodal perception, multi-step execution, tool use, and artifact delivery. However, existing
benchmarks are often tied to specific task types, execution environments, or scoring protocols, limiting their comparability, interpretability, and reliability for deployment decisions. We introduce OmniStepBench, a
benchmark for deployment-aware
agent evaluation across modalities and task complexity. Built on a shared OpenClaw execution environment, OmniStepBench organizes 233 tasks selected and adapted from 22 source
benchmarks into a $2\times3$ workload
matrix defined by input modality and execution form, and evaluates them under a unified contract-based protocol with evidence-based score
auditing. We evaluate 30 commercial API models and 12 locally deployed open-weight models (42 models in total) over 9{,}218 applicable model–task runs, reporting accuracy and
token consumption alongside reference costs for API models. Results show workload-dependent rankings, distinct accuracy–cost trade-offs between text and
multimodal tasks, and competitive but weaker overall performance from local models. These findings support deployment decisions based on workload profiles, deployment mode, and accuracy–cost trade-offs rather than a single aggregate score.
Read the full abstract ▾
All authors Yu Liu, Zhilin Liu, Zhiwei Yang, Shaojie Zhang, Zheyuan Deng, Tingwei Huang, Zhenbo Luo, Lei Jiang, Yanbing Liu, Pei Fu
Artificial Intelligence · Cross-listed from cs.SE
SkillSpec: Intent-Masked Specification Reasoning for Agent Skill Correctness
Yizhuo Zhang, Bo Kang, Yi Yang et al. (+3 more)
Autonomous
agent systems increasingly depend on reusable skill abstractions for consolidating experiential knowledge and domain expertise. These artifacts typically bundle free-form instructions with heterogeneous resources. However, ensuring their correctness remains challenging. Their failure modes transcend conventional code defects to subtle semantic inconsistencies such as intent conflicts, which manifest as silent failures masked by the underlying model. Moreover, skill correctness must be grounded in intended task boundaries and generalizability. We propose SkillSpec, a Hoare-style framework that formulates skill correctness as a specification
reasoning problem. It transforms a heterogeneous skill repository into a unified graph representation that aligns descriptions, instructions and code artifacts. For each node, SkillSpec derives an ExpectSpec from the surrounding declared intent, and infers FactSpecs from encoded behavior under partially disclosed intent. An intent mask regulates access to holistic, lineage, neighborhood, and local views to balance the bias introduced by excessive context against unsupported inference caused by insufficient context. SkillSpec jointly reasons over these views to flag candidate defects, and automatically validates them in an isolated sandbox. On 515 real-world skills from SkillsBench and widely downloaded repositories, SkillSpec identified 763 manually confirmed defects across 239 skills, achieving 61.2% precision. The node-level analysis across multiple model families shows that specification
reasoning is consistently reliable for code nodes, whereas plain-text nodes remain a major bottleneck. Most defects arise at the boundaries between declared intent and implementation, demonstrating that explicit specifications provide a practical foundation for skill quality assurance in real-world
agent ecosystems.
Read the full abstract ▾
All authors Yizhuo Zhang, Bo Kang, Yi Yang, Zhiyu Duan, Zhouteng Ye, Shunkun Yang
Artificial Intelligence · Cross-listed from cs.CV
Image-Scale Robustness and Visual Recognition Performance: A Cross-Architecture Analysis
Anish Monsley Kirupakaran
The sensitivity of visual recognition models to changes in image scale is well established, yet the factors governing this sensitivity across heterogeneous architectures remain unclear. In this work, we investigate whether scale robustness exhibits a common quantitative structure across modern
vision models. We evaluate 20 pretrained ImageNet-1K classifiers spanning seven architectural families, including convolutional, mobile, efficient, and
Transformer-based architectures. By systematically reducing input image scale, we construct scale-accuracy response curves and define a characteristic scale as a compact measure of the onset of substantial recognition degradation. We then examine the relationship between characteristic scale and baseline recognition accuracy, model parameter count, architectural family, and representation stability. A strong inverse association is observed between baseline accuracy and characteristic scale (Pearson r = -0.890, R^2= 0.792, p < 10^-6). This relationship remains stable under bootstrap resampling, leave-one-architecture-out analysis, and leave-one-family-out analysis. In contrast, parameter count provides negligible additional explanatory power after controlling for baseline accuracy (p = 0.80), while architectural family does not provide significant incremental explanatory power. Furthermore, characteristic scale shows essentially no association with representation stability (r = -0.003, p = 0.991). These results indicate that, across the studied models, scale robustness is strongly organized by baseline recognition performance rather than simply by model size, architectural family, or representation stability. The study provides an empirical framework for characterizing scale robustness across
vision architectures and identifies a reproducible accuracy-scale regularity that warrants further theoretical investigation.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from math.CA
A solution to the Erdős Problem #1040
Ioannis Tzachristas
For a compact set $K\subset\mathbb{C}$, let $\vartheta(K)$ be the infimum of the planar areas of the unit lemniscates of all monic
polynomials with zeros in $K$, allowing arbitrary degree and repeated zeros. We prove that $\vartheta(K)=0$ whenever $\operatorname{cap}(K)=1$, with no regularity assumption on $K$. The proof uses a centered harmonic
polynomial that is positive on all but a set of arbitrarily small area in the
polynomial hull of $K$. A
Fourier average of exterior harmonic measures realizes this
polynomial as the logarithmic potential of a signed measure having bounded density with respect to the
equilibrium measure. A positive perturbation and an $L^1$ approximation by empirical measures then produce the required
polynomials. This extends the smooth-boundary result of Krishnapur, Lundberg, and Ramachandran to arbitrary compact sets of capacity one. Together with the capacity-greater-than-one
theorem of Ghosh and Ramachandran and an elementary argument for unbounded sets, it follows that $\vartheta(F)=0$ for every closed infinite set $F\subset\mathbb{C}$ of transfinite diameter at least one, answering the vanishing question in Erdős Problem 1040.
Read the full abstract ▾
Artificial Intelligence
Generator-Independent Runtime Assurance under Partial Observation
Guangxi Wan, Yongbo Xie, Yuqi Liu et al. (+4 more)
Proposal-based controllers—learned policies,
language-model planners, and other black-box generators—are increasingly deployed behind runtime verification gates. We ask when the closed-loop safety guarantee decouples from the generator. The prevailing per-candidate certification pattern does not compose: under retry or best-of-$k$ selection a per-candidate false-admission level $\alpha$ can inflate to $1-(1-\alpha)^{k}$. Our main
theorem shows that simultaneous setwise soundness—certifying a set of admissible proposals containing no nonviable action—is necessary and sufficient for generator-independent admission soundness, the worst case over all generators of executing a nonviable proposal equalling the probability of setwise failure; together with a design-time certificate and a no-bypass rule it is sufficient for contract safety, with violation bound $\Gamma+\sum_t\varepsilon_t+\eta$ invariant under arbitrary, even
adversarial, replacement of the generator. A second
theorem bounds every admission mechanism under partial observation: for a fixed probing and admission policy, if two state hypotheses whose information laws lie within total-variation distance $\delta$ require different safe decisions, then $\abar+\beta+\delta\ge1$. A sequential risk ledger makes the guarantee implementable with time-uniform confidence tubes, and shows that deterministic admission computations concentrate all statistical risk in state estimation. Simplex-style runtime assurance and control-barrier-function filtering are recovered as degenerate cases.
Read the full abstract ▾
All authors Guangxi Wan, Yongbo Xie, Yuqi Liu, Qingwei Dong, Qingxin Li, Hongfei Bai, Peng Zeng
Artificial Intelligence · Cross-listed from cs.CR
Evaluating Deep-Search Agents under Hierarchical Web Evidence Poisoning
Zhongan Bi, Qiwen Wang, Jianrong Jiang et al. (+11 more)
Search-augmented LLM
agents are increasingly used for consumer decisions, making them vulnerable to Generative Engine Optimization (GEO) poisoning. Existing
benchmarks largely measure whether manipulated content is retrieved or endorsed, but do not track whether an
agent verifies suspicious evidence, revises adopted claims, or recovers before producing its final recommendation. We introduce HAE-GEO, a
benchmark that tracks the full trajectory from exposure to recovery under progressively more persuasive Web poisoning.
Agents interact via a multi-turn Search-Scrape interface across three attack levels (L1 direct assertion, L2 contextual camouflage, and L3 apparent corroboration), supported by a controlled corpus of 72,039 clean pages and 770 poisoned pages per level spanning 8 product categories and 154 brands. Evaluation combines deterministic behavioral measures with six semantic rubric dimensions. Evaluating 10
agents, we find three recurring patterns: evidence recognition degrades under the corroboration trap; agentic search improves final resistance without improving evidence recognition or utility; and defense prompting increases verification, yet rarely converts verification into recovery.
Read the full abstract ▾
All authors Zhongan Bi, Qiwen Wang, Jianrong Jiang, Jigang Ding, Wenwen Xiong, Changhua Meng, Xuanang Gao, Kepeng Lin, Changjiang Jiang, Yiang Chen, Huan Yao, Wei Wang, Zhenyu Ma, Wenhui Dong
Artificial Intelligence · Cross-listed from cs.CV
What Does Animal Re-Identification Learn? Linear Biological Concepts and Their Origins in Visual Representations
Robert Nolting, Alexandra Schild, Moritz Weckbecker et al. (+2 more)
Conservation increasingly relies on camera traps that collect more wildlife imagery than experts can manually analyze, making animal re-identification (Re-ID) essential for monitoring individuals and populations. Yet understanding which cues drive model decisions is challenging for ViT-based Re-ID models, whose metric-learning objectives provide no explicit supervision for biological concepts. We ask whether such models nonetheless organize their representations along biologically meaningful axes. Using a DINOv3 backbone fine-tuned for Western lowland gorilla Re-ID with triplet-margin loss, we find that sex and age emerge as linear directions that generalize to held-out individuals, reaching up to 0.91 AUROC and being recoverable from a single image per individual. Activation steering further shows that the sex direction is causally used by the model, flipping a significant fraction of predictions to the opposite sex. Comparing off-the-shelf and fine-tuned backbones shows that Re-ID training does not create these concepts, but relocates them across the network. Finally, data attribution reveals that the representation we find reflects a graded biological axis, is redundantly encoded across the population and shaped by visually ambiguous individuals. Together, these findings show how interpretability can uncover both the biological structure and failure modes of Re-ID representations, providing a step toward auditable computer
vision for wildlife monitoring.
Read the full abstract ▾
All authors Robert Nolting, Alexandra Schild, Moritz Weckbecker, Maximilian Schall, Gerard de Melo
Artificial Intelligence · Cross-listed from cs.CL
Tri-PvP: Exposing Modality Bias in Omni-Modal Large Language Models through Perceptual-Propositional Evidence Conflicts
Yen-Ting Piao, Shu-Yun Chen, Chin-Hui Chu et al. (+4 more)
Omni-modal large
language models (OLLMs) jointly process
vision, audio, and text, yet their modality bias under cross-modal conflict remains underexplored. Existing
benchmarks conflate two distinct forms of evidence within a single modality: perceptual signals (e.g., a photograph or recording of a dog) and propositional signals (e.g., the declarative claim "this is a dog"), such that any measured modality bias is inherently confounded with evidence-form bias, precluding clean attribution to either source. To address this, we introduce Tri-PvP, an 8,000-sample tri-modal conflict
benchmark crossing
vision, audio, and text, where
vision and audio each take perceptual or propositional form. Evaluating five OLLMs, we find robust visual bias across most models and evidence-type conditions. Crucially, we reveal a systematic asymmetry in evidence-form bias: models exhibit a stronger bias toward perceptual signal in
vision but propositional in audio. Further analyses via layer-wise linear probing and contrastive decoding reveal that modality bias is already linearly decodable from early representation layers and can only be partially mitigated, calling for mitigation strategies beyond surface-level interventions.
Read the full abstract ▾
All authors Yen-Ting Piao, Shu-Yun Chen, Chin-Hui Chu, Chun-Wei Chen, Shih-Yun Shan Kuan, Hung-yi Lee, Yun-Nung Chen
Artificial Intelligence · Cross-listed from cs.LG
Memory in Deep Time-Series Models
Minh Hoang Nguyen, Huu Hiep Nguyen, Manh Nguyen et al. (+3 more)
Deep learning for time series has progressed through successive architectural paradigms, from recurrent networks and
transformers to structured state-space models,
retrieval-augmented predictors, foundation models, and tool-using
agents. These developments are typically studied in isolation, organized by architecture or modeling era. We argue that they can instead be viewed through a common question of how does a time-series model retain and access information beyond its immediate input? This question is motivated by a fundamental limitation of conventional time-series modeling: information relevant to a prediction may lie far beyond a feasible input window, while compressing history into a fixed-size state can discard information that may become useful later. We formulate this challenge as a
memory problem and organize existing time-series methods along a spectrum from internal
memory, encoded in parameters and fixed-size states, to external
memory that is addressable, retrievable, and increasingly maintained by
agents. We then develop a unified taxonomy of
memory mechanisms and review three classes of external
memory, including explicit modules,
retrieval augmentation, and agentic stores, under a common framework for what is retained, how it is written and accessed, and how it persists. A cross-cutting analysis maps these mechanisms to time-series tasks and identifies gaps in both methods and evaluation. We conclude by outlining open problems in building
memory systems that can selectively retain, retrieve, revise, and forget information as temporal environments evolve. The result is a framework for studying
memory as a first-class dimension of time series modeling, independent of the underlying backbone.
Read the full abstract ▾
All authors Minh Hoang Nguyen, Huu Hiep Nguyen, Manh Nguyen, Van Dai Do, Dung Nguyen, Hung Le
Artificial Intelligence · Cross-listed from cs.CV
Geometry-Aware Test-Time Learning for Quantitative Spatial Reasoning
Gege Zhang, Shuaicheng Niu, Gang Dai et al. (+2 more)
Quantitative spatial
reasoning in visual-
language models (VLMs) aims to infer spatial distances and directional relationships among objects in 3D space from a 2D image and a natural
language query. Despite recent progress, VLM spatial
reasoning remains brittle under distribution shifts, largely due to the high cost of 3D supervision. As a result, models often produce inconsistent or contradictory predictions when faced with novel object configurations or rephrased spatial queries, revealing a misalignment between learned representations and underlying geometry. To address this, we propose TTL-SR, a geometry-aware Test-Time Learning framework for quantitative Spatial
Reasoning that leverages
geometric consistency constraints and unlabeled test data to adapt models to target domains. Specifically, TTL-SR augments the input query with geometrically coupled auxiliary queries, filters unreliable predictions via adaptive
geometric triggering to construct structured
token-level pseudo-labels, and updates model parameters under a geometry-aware multi-objective loss using only test data. Experimental results demonstrate that TTL-SR significantly boosts spatial
reasoning performance, yielding 6.47% and 9.41% accuracy gains for Qwen3-VL-4B-Instruct and SpatialRGPT-VILA-1.5-8B on Q-Spatial-ScanNet
dataset, respectively.
Read the full abstract ▾
All authors Gege Zhang, Shuaicheng Niu, Gang Dai, Lei Sun, Shuangping Huang
Artificial Intelligence
Agentic Pressure: The Endogenous Entropy of Reliable Autonomy
Hengle Jiang, Ziying Luo and Ke Tang
Achieving reliable autonomy in the wild requires
agents to sustain continuous operations across long-horizon trajectories. However, as
agents navigate these unconstrained settings, they encounter cumulative friction that inherently destabilizes their
alignment. In this paper, we identify a distinct non-
adversarial phenomenon termed Agentic Pressure. We define this as a kinetic force that spontaneously emerges when the cost of compliance conflicts with the imperative of goal achievement. Unlike static jailbreaks, this pressure is endogenous and arises directly from the dynamics of interaction. We propose a theoretical framework that formalizes Agentic Pressure as the ratio between the required work to overcome environmental friction and the remaining capacity of the
agent. Our analysis demonstrates that when this pressure exceeds a critical threshold,
agents exhibit safety drift as a mathematically optimal adaptation. Consequently, they often resort to Instrumental
Hallucination to rationalize rule violations. Empirical experiments validate this framework and show that aligned
agents spontaneously compromise safety to preserve autonomy under high-pressure conditions.
Read the full abstract ▾
Artificial Intelligence
MOAE: Multi-Objective Agent Evolution with Pareto-Preserving Search
Hengle Jiang, Qijun Cai, Ziying Luo et al. (+1 more)
As LLM-based
agents continue to advance, their evaluation has become increasingly multifaceted: a capable
agent must not only achieve high task completion accuracy but also perform well in interaction quality, safety, and efficiency, raising a central question: can these objectives be optimized simultaneously? Existing methods have considered multiple objectives, but many collapse heterogeneous measurements into a fixed scalar score. Such scalarization depends on metric normalization and preference weights and may discard candidates that represent useful deployment trade-offs. We introduce Multi-Objective
Agent Evolution (MOAE), which organizes iterative in-context refinement as a Pareto-preserving evolutionary search over complete
agent rollouts. Given a limited rollout budget, MOAE maintains an empirical archive of non-dominated candidates, uses objective-specific diagnostics to guide offspring generation, and applies constraint-aware selection only at deployment. This separates candidate preservation during search from the preference used to return a final solution. The procedure requires no parameter updates and allows each objective to be replaced by any measurable property, which we instantiate as task performance, trajectory quality, and safety. Experiments on TravelPlanner and AgentDojo show that MOAE consistently improves task performance and trajectory quality while maintaining strong safety under matched rollout budgets. Search-behavior analysis further shows that Pareto preservation expands the attainable objective region and increases the frequency of joint improvement. These results demonstrate the potential of Pareto-preserving in-context
evolution for optimizing multiple
agent properties without committing to a fixed scalarization during search.
Read the full abstract ▾
All authors Hengle Jiang, Qijun Cai, Ziying Luo, Ke Tang
Artificial Intelligence · Cross-listed from cs.LG
DART: Distributional Adversarial Recurrent Training for Algorithm Learning
Hieu Tran Bao, Phung Thanh Dang, Pham Quang Nhat Minh et al. (+1 more)
Recurrent
reasoning models (RRMs) can solve structured problems, achieving easy-to-hard generalization through iterative computation in hidden space. These models are typically trained with instance-level supervision, which becomes increasingly problematic as task difficulty grows: valid solutions occupy a tiny region of the solution space, while invalid solutions proliferate rapidly. We propose Distributional
Adversarial Recurrent Training (DART), a training framework that replaces single-point supervision with a local target distribution around the ground-truth solution and aligns model outputs with this distribution through an
adversarial objective. DART provides a richer learning signal and encourages more stable iterative trajectories toward valid solutions. When evaluated on Maze, Chess, and masked Sudoku with multiple RRMs, including Deep Thinking Systems and Tiny Recursive Models, DART improves solution quality, stability, and robustness under the evaluated distribution shifts. Comparisons with label smoothing, Gaussian softened targets, and progressive training show that DART is not explained by target softening alone and is complementary to training schemes that stabilize long-horizon recurrence. These results identify DART as a promising approach for improving robustness across the evaluated recurrent
reasoning models.
Read the full abstract ▾
All authors Hieu Tran Bao, Phung Thanh Dang, Pham Quang Nhat Minh, Hoang Thanh Tung
Artificial Intelligence · Cross-listed from cs.CL
Beyond Cross-Lingual Transfer: Benchmarking Propagation Boundaries in Multilingual LLM Unlearning
Pengyang Shao, Chuanpeng Lu, Wei Qin et al. (+5 more)
Large
Language Model (LLM) unlearning aims to suppress target knowledge while preserving general capabilities. In
multilingual settings, unlearning must additionally propagate within its intended linguistic scope. However, existing evaluations mainly measure cross-lingual transfer and cannot distinguish insufficient from excessive propagation. We introduce CLLPU (Cross-Lingual and
Language-Bound Protocol for LLM Unlearning), a
multilingual benchmark that formulates this problem through two settings: common-goal forgetting, where target knowledge should be suppressed across all
languages, and
language-conditioned forgetting, where suppression should remain confined to a designated
language. CLLPU combines goal-guided topic pairing, schema-aware relation matching, and dual-anchor
multilingual translation to construct 800 matched knowledge-unit pairs and 72,000 QA instances across ten
languages. Experiments with six representative methods on Llama-3.1-8B-Instruct reveal opposite failure modes: forgetting remains incomplete when universal suppression is required, yet spreads beyond the intended boundary when
language-conditioned confinement is required. We further find that general
multilingual utility can conceal damage to neighbor knowledge. These findings establish propagation control as a central challenge for
multilingual LLM unlearning. We publicly release CLLPU together with its construction pipeline.
Read the full abstract ▾
All authors Pengyang Shao, Chuanpeng Lu, Wei Qin, Yanzheng Jin, Xiaohao Liu, Xi Ai, Kenji Kawaguchi, Richang Hong
Artificial Intelligence · Cross-listed from cs.CY
Intent Drift at SME Scale: Deployment Practice, Not Model Capability, Determines Agentic Compliance
Ilia Voroshilov
We introduce Chain of Intent, a governance framework for agentic AI at small regulated firms, and validate it against a failure it was built to address. Existing agentic governance research assumes enterprise infrastructure that small firms do not have. In a simulated Hong Kong asset manager with 415 synthetic contact records, an
agent performing a routine client-communications task was subjected to ordinary managerial pressure to increase its reach. With its authorised constraints written into its configuration, the
agent held: it identified every ambiguity in the firm's records, cited
privacy legislation it had never been shown, and refused six successive requests, breaching in two of fifteen runs. With the same task, data, pressure and model, but its purpose left unstated as resource-constrained firms routinely leave it, it breached in thirteen of fifteen runs, contacting up to 220 individuals of whom 94 per cent had no demonstrable marketing consent - conduct carrying a maximum of three years' imprisonment under Hong Kong law. Chain of Intent applies four controls requiring no security engineering: a machine-readable purpose, constrained tool access, a scope ledger, and a pre-action check. It eliminated unlawful contact in every run while preserving task completion, and ablation shows each control independently sufficient by a different mechanism. We further show that drift must be measured at two stages -
agents widened their candidate sets in every pressured run while acting on them in roughly one in seven - and that governance applied at the point of intent costs roughly half as much as governance applied at the point of action.
Read the full abstract ▾
Artificial Intelligence
Beyond Final Decisions: A Process-Centric Benchmark for Transparent AI-Assisted Peer Review
Siming Yuan, Xueyi Zhang, Wangze Ni et al. (+7 more)
Peer review is central to quality control in science. However, existing evaluations of AI-assisted peer review mainly focus on the overall quality of generated reviews or the accuracy of final decisions. They therefore provide limited evidence about whether model decisions are supported by sufficient and reliable review evidence. We introduce a process-centric diagnostic
benchmark for AI-assisted peer review. It uses (x,$z_s$,$z_c$,$z_r$,y) to represent the paper content, summary, critique, suggestion, and decision. We convert heterogeneous review records from PeerRead, NLPeer ARR-22, and OpenReview-ICLR into process-aligned data. Our
benchmark uses direct decision prediction from the paper content (Direct) as its baseline. It compares the decision value of Gold-process variables and Predicted-process variables, and conducts stage-level evaluation, chain-consistency evaluation, and interventional sensitivity analysis. Experiments across three
datasets and six models show that Gold-process variables generally have higher decision value. For the main analysis model, the Gold–Predicted gap remains stable across
datasets and random seeds. This gap is also reproduced in most model–
dataset combinations. Although model-generated intermediate review texts show relatively high local consistency across adjacent stages, the final decisions are not consistently supported by the preceding review evidence. Our
benchmark targets AI systems designed to assist rather than replace human reviewers. It provides a transparent and auditable diagnostic tool for evaluating the reliability of their review processes.
Read the full abstract ▾
All authors Siming Yuan, Xueyi Zhang, Wangze Ni, Tianfang Xiao, Shimin Di, Jia Zhu, Zhuoren Jiang, Rong Tan, Lei Chen, Kui Ren
Artificial Intelligence · Cross-listed from cs.CL
Solving versus Verifying: Catching Contradictions in Tax Reasoning Systems
Albert Sadowski and Jarosław A. Chudziak
Large
language models now compute correct tax liabilities on over 90% of well-formed cases in statutory
benchmarks, which makes them candidates for the tax-advisory and compliance systems that consume such an answer directly. Real legal inputs, however, are frequently defective: required facts are missing, or stated facts contradict one another. Accuracy on clean
benchmarks says nothing about how a model behaves then, and a system that computes straight through a defective input returns a confident number with no sign that anything is wrong. This raises two questions: does a model asked to solve a case abstain when the input is defective, and when it does not, can the same model catch the defect when asked instead to verify the input? We study six recent models on SARA-derived tax cases under missing-fact and contradictory-fact perturbations. The strongest models abstain when a fact is missing but compute through injected contradictions, returning the clean-input answer 63-76% of the time with no signal of the conflict; asked instead to verify the same input, they flag most of those contradictions. We wire that verification call into a simple contradiction gate: one extra call that abstains when the model reports a conflict. Across all six models it recovers most of the missed contradiction abstention at a clean-accuracy cost of at most about 5 percentage points, with no training and no external tooling. High accuracy on well-formed inputs is therefore an incomplete measure of reliability, and the detection the solver misses is cheaply recoverable with a single self-check.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
AVSplat: Dense-View Feed-Forward 3D Gaussian Splatting with Assist-View Preconditioning
Muyu Xu, Fangneng Zhan, Yu Wei et al. (+2 more)
Pose-free feed-forward 3D Gaussian Splatting enables novel view synthesis from uncalibrated multi-view images. Although more views should improve performance, existing methods often degrade with dense-view inputs because global aggregation spreads
attention over many
tokens, and naive voxel fusion averages many Gaussians into overly smooth representations. We present AVSplat, a framework that turns additional views into reliable signals for both aggregation and representation. Before global
attention, each view performs a single lightweight interaction with a small set of Assist Views chosen for relevance and diversity, and the cached features provide a focused scene context that stabilizes correspondence. For representation, we use adaptive temperature-aware voxel fusion that sharpens attribution under high occupancy, guided by occupancy and point confidence. Crucially, AVSplat restores positive view scaling where performance remains stable or improves as more input views are added, instead of degrading in the dense-view regime. Ablations show that Assist View Preconditioning is primarily responsible for preventing dense-view degradation, while Occupancy-guided Voxel Fusion contributes most of the single-point image-quality gains.
Read the full abstract ▾
All authors Muyu Xu, Fangneng Zhan, Yu Wei, Hanspeter Pfister, Shijian Lu
Artificial Intelligence · Cross-listed from cs.CR
Versioned Transitive Dependency-Closure Binding and Operation-Time Effect Governance for Agent Skills: ClosureBound
Genliang Zhu and Chu Wang
Agent Skills combine instructions with files, packages, tools, models, and services, so operational identity can exceed a signed directory. Recursive or lazy dependencies may change while root-level evidence remains valid, and different surfaces may reach the same durable effect. We present ClosureBound, a reference monitor that prevents authorization transfer across material changes to this heterogeneous closure. Its resolver commits typed graph nodes and
topology. Each grant binds an exact closure root, effect ceiling, purpose/provenance, validity, and epochs. At durability, it re-resolves closure and state, normalizes the operation into an external-effect IR, and admits it only if a joint witness satisfies every bound. Supported equivalent paths share one ceiling. Assuming complete mediation and discovery, authenticated freshness, sound normalization, cryptographic binding, and authoritative linearization, we establish metadata non-authority, closure determinism, version non-inheritance, effect non-amplification, bound-value freshness, and path invariance. We do not establish program equivalence or remote-service honesty. A provider-free implementation matches 40 frozen lifecycle fixtures; 18 kernel contracts and six mutants cover binding and downgrade cases. Full-profile exploration reaches 84,608 states and 530,752 transitions without a declared invariant violation; six weakened profiles yield witnesses. A lexical audit of 549 public Skills (4,872 unique files) finds that 21 of 526 roots with bundled files name every non-manifest path verbatim, 67 contain links resolving outside their roots, and no root declares a frontmatter dependencies field. These observations motivate conservative closure discovery and define concrete targets for broader runtime, interoperability, efficacy, and production validation.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
STAR-Pro: Stage-Wise Token Adaptive Reduction with Progressive Refinement for Efficient Large Vision-Language Models
Yichen Guo, Tinghao Wang, Qizhe Zhang et al. (+15 more)
Large
vision-
language models (LVLMs) achieve strong
multimodal understanding, but the hundreds to thousands of visual
tokens they process impose substantial computational overhead, motivating training-free visual
token pruning. In this work, we conduct two complementary analyses of visual
token pruning. First, we measure the feature-space coverage of
tokens retained before cross-modal fusion and find that aggressive
pruning discards substantial visual information. Second, we track text-to-visual
attention across decoder layers and find that the visual
tokens considered important change substantially with depth, making one-shot
pruning decisions unreliable. Together, these findings show that effective
pruning should preserve broad visual coverage before fusion and progressively refine the retained
tokens as cross-modal evidence evolves during fusion. We therefore propose
STAR-Pro (STage-Wise Adaptive
Token Reduction with Progressive Refinement), a training-free two-stage framework. Its Adaptive Stage applies pivoted QR to construct an over-budget feature-coverage candidate pool, while its Progressive Stage uses evolving text-to-visual
attention at selected decoder layers to prune a nested survivor set under a target layer-average
token budget. Extensive experiments across seven LVLMs spanning multiple architectures and 18 image and
video benchmarks demonstrate the effectiveness of
STAR-Pro under aggressive
pruning. On LLaVA-
Video-7B,
STAR-Pro reduces visual
tokens by 90.5%, retains 92.7% of baseline performance, and achieves a $2.24\times$ measured inference speedup. Code is available at https://github.com/EasonAI-5589/starpro.
Read the full abstract ▾
All authors Yichen Guo, Tinghao Wang, Qizhe Zhang, Lingbei Meng, Yuan Zhang, Jiajun Cao, Hao Jiang, Chenwei Wu, Jixian Wu, Sixiang Chen, Tao Luo, Hongyang Cheng, Kai Tang, Chenxi Li, Renyuan Li, Xiande Huang, Wenya Wang, Shanghang Zhang
Artificial Intelligence · Cross-listed from cs.CR
Structurally Close, Temporally Distant: Measuring Security Exposure in Long-Horizon LLM Agents
Md Jafrin Hossain and Nur Al Hasan Haldar
Long-horizon LLM
agents interact with untrusted content, persistent
memory, external state, and sensitive tools. Existing analyses often characterize attacks by the number of execution steps between malicious input and a downstream action. We show that temporal remoteness can overstate security separation in stateful
agents. We introduce a provenance-aware execution graph linking
agent events through deterministic state, identifier, and tool provenance, and define influence distance $\DI$ as the shortest structural path from an untrusted source to a sensitive action. We compare it with sequence distance $\DT$, the shortest injection–sink path in the ordered trajectory. Since the influence graph contains every sequence edge, $\DI \leq \DT$; $\Gap=\DT-\DI$ measures the separation hidden by step count. Across 454 injection–sink pairs from 360 long-horizon AgentDojo trajectories over OpenAI's gpt-4o-mini and gpt-4o and Claude's Haiku 4.5 and Sonnet 4.6, $\Gap>0$ for 96.9% of pairs, with a median gap of 9 hops; 91.0% remain decoupled after removing the largest provenance-only edge class. On AgentDojo's banking suite, 33.8% of 231 pairs from 377 trajectories decouple through different provenance mechanisms. Among 274 OpenAI pairs, $\Gap$ does not independently predict attack success after controlling for $\DT$, attack family, and backend ($\beta_{\Gap}=0.066$, $p=.088$). At matched thresholds $k=2,3$, a deterministic $\DI$-based pre-execution gate blocks five attack sinks missed by a sequence-only gate with no additional benign blocking, although the paired gain is not significant ($p=.0625$). Execution structure therefore reveals proximity hidden by step count and can support targeted runtime intervention. We measure candidate influence pathways rather than causal attribution.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
UniRRM: Unified Reasoning Reward Models Across Languages and Evaluation Paradigms
Peng Lai, Yichao Du, Junchao Wu et al. (+6 more)
Reinforcement learning (RL) excels on tasks with verifiable rewards, but in open-ended tasks, the reliability of reward models remains a key challenge. Existing solutions either depend on costly proprietary LLM-as-a-Judge systems or opaque scalar reward models that lack interpretability. Recent works on generative reward models offer a promising alternative, but they remain constrained by static evaluation criteria, fragmented evaluation paradigms, and limited
multilingual support. To address these challenges, we introduce MixReward, a large-scale
multilingual dataset spanning six domains and 103
languages, containing both pairwise and listwise data, and propose UniRRM, a unified
reasoning reward model supporting multiple
languages and evaluation paradigms. UniRRM uses a staged
reasoning chain to dynamically generate task-generic and instruction-specific criteria, enabling fine-grained, input-adaptive judgments while maintaining consistency across
languages. Experiments demonstrate that UniRRM-8B and UniRRM-14B achieve performance close to the state-of-the-art for models of comparable size across multiple
benchmarks, and are effective for unseen evaluation paradigms. In addition, ablation studies validate the reliability and effectiveness of UniRRM.
Read the full abstract ▾
All authors Peng Lai, Yichao Du, Junchao Wu, Weibo Gao, Linan Yue, Longyue Wang, Weihua Luo, Derek F. Wong, Guanhua Chen
Artificial Intelligence · Cross-listed from cs.CR
From Review to Authorization: Key-Isolated Threshold Signing for LLM Agents
Yu Zheng and Qizhi Zhang
Autonomous LLM
agents can turn untrusted content into effectful actions such as payments and permission changes. If the same process interprets this content and controls a reusable signing credential,
prompt injection can cross the judgment boundary and reach execution authority. We present KITA, a review-to-authorization architecture that keeps the user's personal secret signing key and every threshold signing-key share outside all LLM processes. Under threshold signature unforgeability and our system assumptions, compromising the proposer and fewer than t reviewer-signer domains cannot produce a valid authorization for a new action without signing contributions from t distinct domains. Thus, any such authorization includes a share from an uncompromised domain, bound to the canonical action and released only after authenticated reviewer approval. This establishes execution-bound authorization integrity. We implement the complete reviewer-to-executor path with a structured-output LLM adapter and threshold BLS. Six system tests validate quorum gating and message binding at this interface, while cryptographic microbenchmarks measure the online signing path and its scaling behavior.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection
Peng Lai, He Zhu, Zhiwen Ruan et al. (+6 more)
Aligning large
language models with human preferences remains a challenge, primarily due to the critical role of preference data quality in effective
alignment. Existing
datasets are frequently plagued by inherent noise and distribution shifts, which inherently limit model performance. To bridge this gap, we propose AlignDiff, a preference data filtering framework driven by intrinsic model signals. AlignDiff first identifies samples with clear preferences using both positive and inverse signals, then prioritizes the more challenging samples based on the average negative log-likelihood gap, encouraging the model to learn richer information from them. AlignDiff is evaluated on two widely used model families (LLaMA and Qwen) and three
benchmarks widely adopted in the
alignment community (AlpacaEval 2.0, Arena-Hard, and MT-Bench). Across all settings, it consistently outperforms seven strong baselines. We conduct comprehensive ablation studies to validate the effectiveness of AlignDiff, and further show that difficulty-based curriculum learning improves model performance.
Read the full abstract ▾
All authors Peng Lai, He Zhu, Zhiwen Ruan, Dongdong Zhang, Yun Chen, Peng Li, Furu Wei, Yang Liu, Guanhua Chen
Artificial Intelligence
The End of AI Exponentiation: Fluttering Inside and Outside AI Bubble
Victor Kebande
The exponentiation of Artificial intelligence (AI) in the recent past has entered a transformative era that has been driven by the growth in large
language models (LLMs), large-scale compute infrastructures, and autonomous
reasoning systems. However, the rapid acceleration of AI has increasingly shown technological, societal, economic, ethical and infrastructural challenges associated with peak data limitations, rising computational demands, synthetic data recursion, valuation inflation, and societal instability. The traditional scaling paradigms that have powered the modern AI systems are gradually encountering friction in sustaining continuous exponential growth. This paper views “the end of AI exponentiation,” thus exploring how it flutters inside and outside the bubble, where instability emerges within the AI ecosystem through compute and data-center races, speculative investments, and the rat-race toward superintelligence, and outside the ecosystem through labor disruption, governance concerns, public uncertainty, and geopolitical acceleration surrounding future intelligent systems and infrastructures globally.
Read the full abstract ▾
Artificial Intelligence
Multimodal Resource-Exhaustion Attacks on Vision-Language Models via Joint Pixel-Prompt Optimization
Zhaoxiong Ni, Yatie Xiao, Chi-Man Pun et al. (+3 more)
Resource-exhaustion attacks against autoregressive
vision-
language models (VLMs) typically assume unimodal threat models, treating the image branch as the primary optimization surface while holding user-visible
prompts fixed. Even recent loop-centric variants remain confined to this single-channel paradigm, leaving the exploitation of availability unexplored as a cross-modal optimization problem over jointly controllable input surfaces. We introduce Joint Pixel-
Prompt Optimization (JPPO), the first compound
adversarial framework elevating the visible
prompt to a first-class
adversarial variable alongside image perturbations. Under a restricted joint-input threat model, JPPO performs coupled, stagewise optimization over both the pixel and
prompt surfaces. This produces synergistic cost amplification, mechanistically distinct from loop-dependent failures, exhibiting negligible loop incidence in our experiments. Evaluating five open-source VLM families on MS COCO and ImageNet under an 8/255 infinity-norm budget, JPPO achieves over 4.6x latency and 5.3x energy amplification on Qwen2.5-VL-7B, and over 36.6x latency with 32.7x energy amplification on BLIP-2. This represents the strongest cost amplification among directly compared baselines while requiring substantially fewer optimization iterations. Ablations confirm this amplification arises from
multimodal coordination rather than
prompt length or isolated modalities. These findings reveal structural blind spots in current VLM serving defenses, motivating cost-aware robustness evaluation as a first-class security requirement for
multimodal deployments.
Read the full abstract ▾
All authors Zhaoxiong Ni, Yatie Xiao, Chi-Man Pun, Fei Peng, Qingxiao Guan, Keke Tang
Artificial Intelligence · Cross-listed from cs.CL
What if LLMs Ate Their Words: Causal History Effects in Multi-Turn Interaction
Jinnan Li, Zheren Fu, Yue Wang et al. (+3 more)
Multi-turn interaction creates a feedback process in which an LLM's previous responses become context for later behavior. Prior work shows substantial multi-turn degradation and that assistant-generated history can affect later behavior. However, it remains unclear how these effects manifest across models, tasks, turns, and inside a model. We study these gaps across six task families and five models. Degradation from fully specified single-turn input (FULL) to progressively revealed multi-turn interaction (SHARDED) is clearly task- and model-dependent, and stronger one-shot performance does not imply greater interaction robustness. We then retrospectively analyze completed SHARDED conversations by replaying the user messages already observed in each trajectory while editing only assistant-generated history. Replacing prior assistant responses with neutral content (termed neutralization) changes downstream min-max normalized performance by +.027 across 2,973 trajectories. On a prespecified length-controlled subset, short and length-matched neutralization yield nearly identical effects (+.069 versus +.068), showing that simple context shortening is insufficient to explain the effect of history editing. Turn Surgery further intervenes on one assistant turn at a time. Among 237 selected degraded trajectories, 63.7% contain at least one beneficial intervention, while most tested positions remain unchanged; for binary tasks, 48.4% admit a fail-to-success reversal. An open-weight case study links behaviorally consequential history changes to measurable downstream state differences, but finds task-dependent rather than universal internal signatures. Overall, assistant-generated history has active but selective effects on multi-turn performance, motivating selective rather than uniform history management.
Read the full abstract ▾
All authors Jinnan Li, Zheren Fu, Yue Wang, Jinzhe Li, Yuan Wu, Yi Chang
Artificial Intelligence · Cross-listed from cs.LG
Grounded and Faithful P&ID Reasoning: Constraining Vision-Language Models with Recovered Evidence Graphs
Prathamesh Gadekar, Sagar Srinivas Sakhinana and Venkataramana Runkana
Piping and Instrumentation Diagrams (P&IDs) are the authoritative maps of process plants: isolation, maintenance, and HAZOP decisions depend on what connects to what.
Vision-
language models describe these sheets fluently, yet they often invent or miss process connections—and an invented or missed link can reverse an isolation or reachability call, so a plant decision cannot trust a fluent answer that was never checked against the linework. We instead recover an explicit graph of the drawing—its symbols, the process connections between them, and the tags that name them—and then require the model to answer only by querying that graph through seven read-only operators, so a
topology claim is returned only when it cites the query results that support it. On TopoPID-VQA, a new suite of 3000
topology questions over these sheets, Graph-Grounded Harness (Ours) raises exact match accuracy from 36.7–41.3% under image-only prompting to 74.3–76.0% for Qwen3-VL-4B, Qwen3-VL-8B, and Gemma-4-E4B. It does so on an imperfect substrate: on Digitize-PID
dataset the recovered graph scores F1 0.742 on exact process connections, and 0.801 once symbols and tags are pooled in. The residual errors track that gap—grounding pays off where the recovered graph is right, and perception error still breaks
topology questions where it is not.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
FACT: A Forensic Agent with Compiled Tool-Use Trajectories for AI-Generated Image Detection
Jiaoyang Chen, Bin Hu, Jingyu Hu et al. (+3 more)
AI-generated image detection is increasingly open-world: new image generators produce highly realistic images that make visual artifacts harder to identify. Existing detectors usually rely on a fixed set of forensic cues, so a detector that works well for one generator family may fail on another. We introduce FACT (Forensic
Agent with Compiled Tool-use Trajectories), which learns an image-conditioned tool-use policy for forensic analysis. Instead of applying a fixed detector, FACT decides which forensic tools to call, interprets the returned evidence, and stops when sufficient evidence has been collected. FACT follows an Evolve–Distill–Refine pipeline: it evolves an execution-verified forensic skill, compiles the skill into action–observation tool-use trajectories, distills them into a compact
agent, and refines the policy with cost-aware GRPO. Across two internal and four public
benchmarks, FACT achieves the best performance among all compared methods, including on recent unseen generators, deepfakes, and manipulated images.
Read the full abstract ▾
All authors Jiaoyang Chen, Bin Hu, Jingyu Hu, Kun Zhou, Qin Zhang, Zhengzhe Liu