Artificial Intelligence · Cross-listed from cs.SE
SWE-Test: Benchmarking LLM Vulnerability Discovery via Input Prediction
Yuanxiang Shi, Jiayi Lin, Xuanyong Lin et al. (+8 more)
Vulnerability discovery is becoming an important ability of large
language model (LLM)
agents:
agents that silently miss real defects leave critical software exposed. Rigorously measuring this ability is therefore urgent, but existing
benchmarks are gameable through data contamination, score recall against an unknowable vulnerability set, often rely on synthetic bugs, and report a single end-to-end verdict that cannot localize where an
agent fails. Vulnerability discovery is a composite ability: an
agent must comprehend source code, infer input constraints, construct inputs, execute them, and iteratively correct from feedback. We recast its measurement as an input-prediction task with a closed, deterministic ground truth: using coverage-guided fuzzing, we mine deep target branches in real-world C/C++ programs and ask an
agent to predict an input that drives execution to a given branch. This decomposes discovery into three task modes over 22 real-world C/C++ programs spanning 15 domains. Open-loop and Feedback-enabled share 60 fixed-target task instances across 16 of these codebases (13 domains), testing input construction without and with a distance oracle to isolate code comprehension from feedback-driven correction. Online Arena instead removes the predefined target and scores path exploration by coverage gain on a separate, partially overlapping pool of 11 programs;
agents collectively confirmed 13 distinct bugs across six programs. Evaluating 15 default-effort model-scaffold configurations, the best reaches only 55.0% pass rate in the Feedback-enabled mode, and the mean across seven paired Claude Code configurations is 36.4% with feedback versus 19.3% without. Decomposing failures, we find constraint inference, not
navigation, is the dominant bottleneck. We release SWE-Test with a turnkey evaluation environment.
Read the full abstract ▾
All authors Yuanxiang Shi, Jiayi Lin, Xuanyong Lin, Liangcai Su, Yeheng Duan, Wei Wang, Qi Han, Bing Zhao, Wei Hu, Xander Xu, Chenxiong Qian
Artificial Intelligence · Cross-listed from cs.CR
Scratchy: Visual-Scratchpad Multimodal Reasoning for Cryptographic Proof Generation in EasyCrypt
Yupeng Ren, Zhaoxuan Li and Rui Zhang
Large
language models (LLMs) have recently made substantial progress in formal proof generation, yet presenting distinctive challenges in cryptographic area. Computational security arguments posit that a valid proof must coordinate probability,
adversarial games, invariants, assumptions and bounds, which can be provided by a machine-checked framework named EasyCrypt. Although all objects may appear in available context, LLMs still struggle because proof-theoretic dependencies are typically implicit in a linear representation and distributed across multiple programs. So, this paper presents Scratchy, a visual-scratchpad approach that exposes these dependencies for
multimodal generation. Given the natural-
language security description, with formal context and target propositions, the proof objects can be normalized into a typed proof-relation graph. Then a structure-preserving visual compiler transforms the graph into the formula-rich visual proof state that guides a
multimodal model in generating the EasyCrypt proof. Also, the Scratchy-eval, a 114-task
dataset derived from reliable official EasyCrypt files, has been introduced. It contains 64 security-form proof generations and 50 multiple-choice knowledge tests. After a series of evaluations, covering semantic grounding, relational invariants, and game reductions, classical LLMs like GPT-5.6-Sol and Claude-Opus-5 have gained a clear advantage from Scratchy's structured visual proof states. This contrast suggests that explicit proof structure can make the improvement and
multimodal proof-state representation as a promising direction for computer-aided cryptography.
Read the full abstract ▾
Artificial Intelligence
Predicting Wind Turbine Power Using Machine Learning and Weather Forecasts
Khivishta Boodhoo, Isaac Triguero, Josh Plumbly et al. (+2 more)
Offshore wind turbines are widely used to generate renewable energy, but their maintenance can result in decreased efficiency due to forced shutdowns. Accurate wind turbine power predictions can identify periods of low power that would be ideal for scheduling maintenance. However, the effects of data volume, feature selection, and data preprocessing on the performance of such power prediction models have not been thoroughly studied. Besides, current models have limited transferability between different wind turbines. Therefore, this study developed a baseline Linear Regression for performance comparison with a more complex Artificial Neural Network model to predict the power output of a wind turbine, using weather conditions only to enhance applicability. A range of data preprocessing techniques were studied, and models were trained on one month and one year of data to determine the effects of data preprocessing and volume on model performance. Feature selection was explored using a Random Forest Regressor. The best results from the different models showed that the Artificial Neural Network models provided the highest accuracy, with an R2 score of 0.98 and a low Mean Absolute Error of 194, when compared with the baseline model (R2 score of 0.94 and Mean Absolute Error of 441). The model performance is comparable to the range of results in past studies, with the advantage that the proposed method leverages a separate weather
dataset from a nearby weather station, enabling future applications for similar wind turbines in different locations. The Artificial Neural Network model was then used to identify 4-h periods of low power predictions over 2 months (simulating application for future periods), providing power output savings of approximately 2000 kW for each maintenance event.
Read the full abstract ▾
All authors Khivishta Boodhoo, Isaac Triguero, Josh Plumbly, Bruce Nicolson, Nicholas Watson
Artificial Intelligence
SCIRIGOR:Evaluating Open-Ended Scientific Analysis Beyond Final Scores
Bowen Liu, Shuo Nie, Bodong Du et al. (+1 more)
Scientific coding
agents produce interdependent code, results, figures, and claims, yet evaluating final outputs alone does not establish whether their conclusions are scientifically supported. We formulate evidence-grounded
multimodal scientific analysis, requiring
agents to produce executable analyses and claims supported by results and visualizations from the same run. We introduce SciRIGOR, an evaluation framework and
benchmark comprising 100 cases from scientific articles across six domains and 17 subfields. The framework reconstructs typed evidence graphs, separates artifact fidelity from relational validity, and scores complete claim-support paths while localizing the earliest unsupported relation. Source-grounded alternative paths accommodate scientifically equivalent analyses and visualizations. We evaluate 11
agent/model configurations. On full-
benchmark runs, claims agree with faithful and unfaithful results at nearly identical rates (91.8% versus 91.0%). Yet no system exceeds 62.6% on the soft evidence-chain score or 18.0% strict whole-chain success. These findings show that internal coherence does not establish scientific correctness: evaluation must verify support along the complete data-to-claim path.
Read the full abstract ▾
All authors Bowen Liu, Shuo Nie, Bodong Du, Xiaomeng Li
Artificial Intelligence
Customer Relationship Intelligence: Integrating CRM and MDM for Enhanced Customer Engagement
Tejasvi c. Addagada
This study examines how Customer Relationship Management (CRM), Master Data Management (MDM), and Customer Knowledge Management (CKM) jointly constitute a Customer Relationship Intelligence (CRI) framework for enhanced Customer Engagement (CE). A cross-sectional survey of 100 participants across retail, healthcare, IT, and telecommunications sectors was analysed using Spearman rho correlation and ordinal logistic regression (IBM SPSS). Bivariate correlations were weak and non-significant (r<0.19, p>0.06). Regression identified CRM (beta=0.717, p=0.002) and CKM (beta=0.581, p=0.009) as significant positive predictors of CE; MDM showed a positive but non-significant direct effect (beta=0.346, p=0.071). The model explained 20.5% of CE variance (Nagelkerke R^2=0.205). Parallel mediation analysis (Hayes PROCESS Model 4, 5,000 bootstrap samples) found no significant indirect effects of MDM on CE via CRM (IE=0.021, 95% BC CI [-0.072, 0.121]) or CKM (IE=0.032, 95% BC CI [-0.061, 0.126]); Hypothesis H4 was not supported. CRM and CKM emerge as the principal drivers of CE within the CRI framework, while MDM functions as a foundational data quality enabler whose strategic value is realised through its enabling effect on CRM execution and knowledge management. Findings should be treated as exploratory given the sample size and cross-sectional design. Future research should replicate with larger sector-specific samples and longitudinal designs, particularly in regulated BFSI contexts where MDM architecture is shaped by data governance mandates.
Read the full abstract ▾
Artificial Intelligence
MVFA: A Multi-View Text-Guided Multimodal Fusion LLM Adapter for Sentiment Analysis and Emotion Recognition
Pengfei Shao, Jisheng Dang, Jiawen Fang et al. (+7 more)
Multimodal sentiment analysis and
emotion recognition in conversations demand effective modeling of heterogeneous interactions across textual, acoustic, and visual modalities. Although large
language models (LLMs) offer powerful
language understanding, adapting them to
multimodal affective computing remains challenging: full-model fine-tuning is computationally prohibitive, while many existing lightweight adapters fail to preserve rich textual cues during cross-modal fusion. To address these limitations, we propose the multi-view text-guided
multimodal fusion adapter (MVFA), a parameter-efficient framework that augments frozen LLMs with strong
multimodal reasoning capability. MVFA first constructs complementary text views via max pooling, mean pooling, and
attention pooling; these views then guide cross-modal interactions with audio and visual features. The fused
multimodal representations are subsequently compressed into a compact set of learnable pseudo-
tokens through an Enhanced Q-Former Fusion Module. Using ChatGLM3-6B-base as the primary backbone, we further validate MVFA on LLaMA2-7B and Qwen3-8B to examine its portability across multiple frozen LLM backbones. MVFA is evaluated on three challenging
datasets: CH-SIMS V2.0, MELD, and CHERMA. Experimental results demonstrate that MVFA achieves state-of-the-art performance on key metrics while updating only a small fraction of parameters. Specifically, it attains 84.62% Acc2 and 84.59% F1 on CH-SIMS V2.0, 67.36% Acc and 66.03% WF1 on MELD, and 74.66% Acc on CHERMA. These findings establish multi-view text-guided fusion as an effective and scalable paradigm for parameter-efficient
multimodal LLM adaptation in affective computing. The code is publicly available at https://github.com/Overwhelm1208/MVFA.
Read the full abstract ▾
All authors Pengfei Shao, Jisheng Dang, Jiawen Fang, Ning Liu, Wencan Zhang, Bimei Wang, Jingwen Zhao, Jianhuang Lai, Qi Tian, Tat-Seng Chua
Artificial Intelligence · Cross-listed from cs.DC
Diamond Agent: Agentic Control of Federated HPC Resources as a Service
Haotian Xie, Junlin Chen, Mingkai Zheng et al. (+9 more)
Efficiently aggregating and orchestrating computing power across heterogeneous clusters for HPC workflows faces four practical challenges: preserving workflow context across independently administered clusters, moving large
datasets between sites,
reasoning about site-specific environments and scheduler policies, and exploiting live queue and resource states for efficient task scheduling. To this end, we design Diamond
Agent, an agentic system that enables intelligent execution of HPC workflows across heterogeneous clusters with typed skills as the interface. Diamond
Agent provides an
agent-facing workspace and skills that unify cross-site resource discovery, resource specification, data movement, task execution, and result
retrieval. A centralized Diamond
Agent instance can operate multiple supercomputers without being deployed separately on each login node. Diamond
Agent translates high-level
agent actions into valid site-specific executions, moves data through Globus Transfer, and uses live system capability and queue information to select feasible placements. Its event-driven continuation mechanism decouples
agent actions from long-running batch jobs: persistent services monitor remote execution and resume the
agent only when a result or decision-relevant event is available. We experiment with 27 hours of telemetry and 19 matched multi-site submission rounds comprising 83 jobs across four production supercomputers. Compared with a fixed-site baseline, Diamond
Agent reduces the median additional completion time relative to the fastest observed placement from 42 seconds to 4 seconds, a 10.5x reduction.
Read the full abstract ▾
All authors Haotian Xie, Junlin Chen, Mingkai Zheng, Yifan Zhu, Minu Mathew, Max Burnette, Yadu Babuji, Volodymyr Kindratenko, Shivaram Venkataraman, Kyle Chard, Ian Foster, Zhao Zhang
Artificial Intelligence · Cross-listed from cs.LG
Decision-Aware Suffix Prediction and Reasoning of Business Processes
Henryk Mustroph and Stefanie Rinderle-Ma
Suffix prediction forecasts the remaining sequence of events of a running case until completion. Most approaches rely on neural networks trained on event logs, which, on average, perform well but struggle with short prefixes or targets belonging to a rare process variant. In such scenarios, the correct path may cross multiple branching decisions, determined primarily by case- and event-level attributes, a signal that NN-based suffix prediction models tend to underweight because they may heavily weight (dense) event labels. Decision mining extracts rules for such decisions from the event log, but has so far been applied only to post-hoc and what-if analysis, not suffix prediction. We therefore extend suffix prediction with decision mining, introducing a decision-aware suffix prediction framework, a neuro-symbolic approach that enables
reasoning about predicted events via mined decision rules. Experiments on three of four event logs and three suffix predictors show that the framework can improve suffix prediction, especially for short prefixes but also for rare process variants, and adds intrinsic interpretability.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
Multiple Myeloma Lesion Segmentation on Whole-Body Diffusion-Weighted Imaging via Efficient Anatomical Anticipation and Multimodal Confirmation
Mengmeng Zhang, Shengqian Huang, Junde Zhou et al. (+12 more)
Whole-body
diffusion-weighted imaging (WB-DWI) is widely used for multiple myeloma (MM) assessment, yet automated lesion
segmentation remains challenging due to limited anatomical delineation and the low specificity of marrow hyperintensity. Existing studies have introduced bone region-of-interest (ROI) information and apparent
diffusion coefficient (ADC) maps to mitigate these ambiguities, but practical limitations remain. Bone ROI construction often relies on costly manual annotation, image registration, or dedicated bone models, while ADC is usually incorporated only through simple channel fusion, limiting its ability to provide complementary structural and lesion-discriminative cues. To address these limitations, we propose a two-stage framework for MM lesion
segmentation on WB-DWI. In the first stage, we train a bone ROI generation model from ADC images without dedicated bone labels, providing an efficient and practical anatomical prior for lesion analysis. In the second stage, we propose Anatomy-guided
Multimodal U-Net (AMU-Net), which leverages ADC in a manner consistent with clinical lesion assessment rather than treating it as a generic auxiliary modality. Extensive experiments demonstrate the effectiveness and practicality of the proposed method. It achieves the best overall performance among the evaluated methods, with a mean Dice score of 76.2%.
Read the full abstract ▾
All authors Mengmeng Zhang, Shengqian Huang, Junde Zhou, Xiaoping Wu, Hao Luog, Jing Wanga, Yicheng Sun, Jiao Li, Haibo Zhang, Sheng Xie, Fan Wangg, Qin Wangc, Huadan Xue, Yisheng Lv, Fei-yue Wang
Artificial Intelligence · Cross-listed from cs.LG
All for 1-Bit: Towards Genuine 1-Bit Post-Training Quantization for LLMs
Zhixiong Zhao, Zukang Xu, Guangyu Sun et al. (+2 more)
Large
language models (LLMs) have achieved remarkable progress, yet their massive storage and
memory-bandwidth demands still hinder efficient deployment. Weight binarization is a promising solution, but existing binarization-based post-training quantization (PTQ) methods usually far exceed the nominal 1-bit storage target due to hidden overhead. To address this gap, we propose All for 1-Bit (AF1), a genuine 1-bit PTQ framework for LLMs. AF1 comprises two complementary components: (1) Null-space-Aware Binary Factorization (NABF) for improving binary reconstruction through Hessian-aware surrogate reparameterization, null-space-aware binary factorization, and scale-only global reconstruction; and (2) Hierarchical Shapley Allocation (HiSA) for assigning structural capacity using hierarchical Shapley sensitivity. Together, they preserve model accuracy under a strict 1.0-BPW budget in the PTQ setting. Experiments on LLaMA, Qwen, and Gemma families show that AF1 consistently outperforms existing binarization-based PTQ methods in perplexity and zero-shot accuracy. Compared with BF16, AF1 achieves an average 2.5 times inference speedup and over 90%
memory reduction across evaluated models, providing a practical path toward deployable genuine 1-bit compression for LLMs. The code for reproducibility is available at https://github.com/Kishon-zzx/AF1.
Read the full abstract ▾
All authors Zhixiong Zhao, Zukang Xu, Guangyu Sun, Lifeng Liu, Dawei Yang
Artificial Intelligence · Cross-listed from cs.AR
From Splats to Silicon: Rethinking Computational Efficiency of 3DGS
Minnan Pei, Qiwei Dong, Yihan Zhou et al. (+7 more)
3D Gaussian splatting (3DGS) represents scenes with explicit primitives and supports real-time novel-view synthesis, yet its system efficiency varies substantially across scenes, viewpoints, rendering paths, and platform constraints. Existing studies pursue efficiency through representation and
algorithm design, GPU runtime optimization, and architectural support, but their reported gains correspond to different points along the rendering and update paths. Connecting these indicators to end-to-end system benefit requires tracing how each optimization changes Gaussian selection, screen-space work, data movement, and stage or frame time. We therefore use a workload-centric framework to connect representation and
algorithm research, GPU runtimes, and hardware architectures and to identify recurring workload patterns. We complement literature analysis with reproduced measurements and controlled GPU profiling of selected implementations, relating workload counts to stage time and
memory traffic. Together, these comparisons show that system gains depend on workload reductions reaching downstream execution, granularity matching each stage, and the cost of data transfers, synchronization, and cached results, gradients, and optimizer data. Building on these findings, we discuss more consistent evaluation under rendering-quality constraints and identify key directions for future system design.
Read the full abstract ▾
All authors Minnan Pei, Qiwei Dong, Yihan Zhou, Gang Li, Yuchen Zhu, Wenju Zhao, Zhongtian Long, Siting Wang, Peisong Wang, Jian Cheng
Artificial Intelligence · Cross-listed from cs.IR
ExpertLens: Visualizing Embedding Spaces for Post-Hoc Explainability in MoE Enhanced Retrievers
Effrosyni Sokli, Isaac Roberts, Alexander Schulz et al. (+2 more)
Neural models, including dense retrievers, have been widely adopted in Information
Retrieval (IR), often delivering state-of-the-art performance. Despite their effectiveness, these models operate as black boxes, limiting the interpretability of their ranking decisions. Existing post-hoc explainability methods for neural rankers primarily focus on feature-level attributions, which can be insufficient to capture the complexity of learned
embedding spaces. In this work, we propose ExpertLens, a post-hoc explainability framework for Mixture-of-Experts (MoE)-enhanced dense retrievers that shifts focus from local scalar feature importance to representation-level global interpretability. ExpertLens leverages discriminative
embedding space visualizations jointly with automatically extracted Concept Activation Vectors to reveal how expert routing drives
embedding space formulation and
retrieval effectiveness. Experiments across five IR
benchmarks and two MoE-enhanced dense retrievers show that expert routing consistently improves
embedding space structure, positioning queries and their relevant documents into better-defined
geometric neighborhoods. Analysis of expert subspaces further reveals general-purpose dominant experts, along with minority experts exhibiting distinct linguistic specialization, with subspaces arranged according to multi-semantic concept similarity. Our code is publicly available.
Read the full abstract ▾
All authors Effrosyni Sokli, Isaac Roberts, Alexander Schulz, Barbara Hammer, Gabriella Pasi
Artificial Intelligence · Cross-listed from cs.CR
SCRIPTIOC-BENCH: A Benchmark for Recognizing Actionable Threat Intelligence from Script-Based Malware using LLMs
Hanna Kim, Jian Cui, Minkyoo Song et al. (+4 more)
Script-based malware remains a prevalent attack technique. These scripts often contain indicators of compromise (IOCs) that provide actionable threat intelligence. However, statically recovering such indicators is challenging, as relevant values may be dispersed or transformed within code. Although large
language models (LLMs) have shown promise in security analysis, their ability to recover IOCs from malicious scripts remains underexplored. We present SCRIPTIOC-BENCH, a
benchmark for measuring static IOC extraction capability on real-world malicious scripts. The
benchmark comprises 634 manually verified JavaScript, PowerShell, and VBScript malware samples covering four IOC types (URLs, domains, IP addresses, and filesystem artifacts). We further stratify ground-truth IOCs by recovery level, distinguishing directly exposed indicators from those requiring decoding or reconstruction. Using this
benchmark, we evaluate a broad range of proprietary and open-weight LLMs and show that IOC recovery without execution remains challenging across model scales: the strongest model reaches only 65.4 F1. To characterize how recovery fails, we introduce a false-positive taxonomy and use it to compare the error profiles of the evaluated models. We further study two mitigations on a small open-weight model, deterministic string utilities and task-specific adaptation, finding that they provide complementary recovery gains, raise precision, and shift errors toward sample-grounded mismatches.
Read the full abstract ▾
All authors Hanna Kim, Jian Cui, Minkyoo Song, Hwanjo Heo, Seungwon Shin, Kimin Lee, Xiaojing Liao
Artificial Intelligence · Cross-listed from cs.CL
What the Window Does Not Contain: Auditing Provenance in a Document-Grounded Instability Benchmark
Seyed Mosayeb Alam
Ask a
language model the same question about the same document twenty times, and it sometimes returns two different answers. We built Probity, a
benchmark of 60 tasks and 470 items from real venture-financing filings, to measure how often this happens. Then we audited our own corpus and found a defect any excerpt-built
benchmark can carry: items whose evidence is missing from the window of text the model is shown. The audit flags 36 items and separates two failures a single flag would conflate: evidence genuinely absent from the window and answers that must be computed from numbers the window does supply. Flagged items change their answers far more often, wobbling at 0.255 against 0.087 on the 427 clean items, and excluding them cuts apparent cross-model agreement by about a fifth. Before testing whether the missing evidence explains the instability, we registered a prediction: re-cut each window to hold its evidence, and instability should fall below a set threshold. It failed: the repair moved wobble by 0.058, with an interval containing zero. We report the association as correlational. Almost all measurements sit where instability cannot show, which bounds what a corpus built for accuracy can say about stability. We release the corpus, all 112,800 raw responses, and the audit as a runnable check for any document
benchmark.
Read the full abstract ▾
Artificial Intelligence
IIns-VAE+: A Robust Transfer Learning Framework for Environmental Identification in Wireless Sensing
Yuxiao Li, Keke Hu, Bobai Zhao et al. (+2 more)
Environmental identification in wireless sensing is essential for 6G integrated sensing and communication (
ISAC) systems to achieve reliable situational awareness. However, deep learning (DL) models for this task often fail to generalize under domain shift across diverse environments. While the Inter-Instance Variational Auto-encoder (IIns-VAE) learns features of rich representation, its neural classifier remains vulnerable to these distribution changes. In this paper, we propose IIns-VAE+, a hybrid model that combines the IIns-VAE framework with Minimax Risk Classifiers (MRC) to improve adaptability in transfer learning scenarios. We use real-world
datasets to evaluate our framework across three transfer learning scenarios, including general to specific room environments, high to low label resolutions, and mixed to specific environments. The experimental results indicate that IIns-VAE+ significantly outperforms baselines, demonstrating its critical value in building adaptable and robust perceptive networks in future 6G systems.
Read the full abstract ▾
All authors Yuxiao Li, Keke Hu, Bobai Zhao, Santiago Mazuelas, Yuan Shen
Artificial Intelligence
Substrate-Portable Execution for Production LLM Workflows
Tarun Gopinath, Atul Kulkarni, Vijay Rajakumar et al. (+2 more)
Production LLM
agents execute tool-calling loops,
retrieval chains, and compositional workflows in multiple modes, yet execution semantics are often coupled to one runtime. We encountered this portability problem in Rufus, a conversational AI assistant with a large tool catalog that serves millions of Amazon customers. Rufus supports real-time serving, asynchronous background tasks, and high-volume batch workloads such as evaluation and content pregeneration. Each mode has distinct service-level objectives and typically uses a separate runtime. Reusing streaming orchestration makes asynchronous and batch workloads blocking and prevents use of batch inference APIs, which offer a 50 percent discount at published prices. We present a binding-adaptive
agent execution platform that separates workflow definition from execution substrate. Developers define a workflow once as a typed dataflow graph. The platform compiles the graph to in-process streaming for real-time serving, durable AWS SWF orchestration for asynchronous execution, or distributed Apache Flink stream processing for batch inference. No workflow code changes are required. LLM inference is represented as a suspendable graph node whose behavior depends on the substrate: streaming delivery online, durable retry asynchronously, and batched submission offline. We validated dozens of production
agent configurations across five orchestration patterns: single-inference RAG, iterative ReAct, compositional PreAct, conditional routing, and multi-
agent deep research. Across all three bindings, we found no detectable difference in output quality. Batch execution reduced per-query inference cost in line with published batch API pricing while operating alongside the streaming path at production scale.
Read the full abstract ▾
All authors Tarun Gopinath, Atul Kulkarni, Vijay Rajakumar, Shrikar Katti, Parthasarathy Govindarajen
Artificial Intelligence · Cross-listed from math.RT
PAGR: Proof-Carrying Algebraic-Geometric Retrieval: A Quiver-, Provenance-, and Sheaf-Theoretic Framework for Grounded LLM Retrieval
Xingting Wang and Min Wu
Retrieval-augmented generation is usually formulated as a statistical information-
retrieval problem. Graph-based variants add relational structure, but the mathematical status of that structure is often left underspecified. Three distinct questions tend to be conflated: which statements are certified as knowledge, which latent representations are useful for
retrieval, and which multi-hop compositions are semantically admissible. We propose Proof-Carrying Algebraic-
Geometric Retrieval (PAGR), a framework that separates these questions mathematically. Its symbolic layer is a many-sorted relational theory generated by a typed quiver, path equations, and positive Horn inclusions. A quiver representation assigns inner-product spaces to entity types and linear operators to relations. A cellular sheaf measures local-to-global consistency. Semiring provenance records derivations and supports machine-checkable certificates. The central principle is epistemic separation: learned geometry may rank and organize evidence, but cannot promote a hypothesis to certified ground truth. We show the certification criterion is invariant under arbitrary replacement of learned components. Further results include a conditional completeness bound, identification of the isometry group as the relevant
symmetry for residual-based
retrieval, a cohomological consistency diagnostic, and a bounded-bisimulation index for admissible-path expansion. PAGR is a mathematical architecture for separating where a system should look from what it is allowed to treat as knowledge.
Read the full abstract ▾
Artificial Intelligence
CWF: A Collaborative Writing Framework for Personalized and Reliable Popular Science Writing
Ruibiao Fu, Di Tang, Yunlong Yang et al. (+7 more)
We introduce Personalized and Reliable Popular Science Writing, a novel task that requires adapting scientific explanations to audiences with different cognitive levels while preserving factual accuracy. However, improving personalization often introduces simplifications that increase the risk of
hallucination and factual distortion. To address these challenges, we first construct a
dataset of 39,134 entries and a reader-centric Personalized Science Communication
Benchmark (PSCB) that jointly evaluates audience adaptation and factual accuracy. To reduce data and computational requirements while improving generalization across domains and audiences, we introduce DA-MoE, which explicitly decouples audience adaptation from domain knowledge through separate modeling. To enable robust verification and revision in evidence-scarce scenarios, a multi-
agent fact-checking mechanism that augments limited evidence with role-specific
agent debate and propagates confidence over a graph is proposed. Experiments on PSCB show that our approach achieves state-of-the-art performance. Our code is open-sourced at https://github.com/DPInnovationWorks/CWF.
Read the full abstract ▾
All authors Ruibiao Fu, Di Tang, Yunlong Yang, Ran Wang, Sicheng Lu, Peixuan Wu, Xiaoyu Fan, Jiacheng Ma, HaoZhe Luo, Yang Xiao
Artificial Intelligence
SAP: State-Guided Data Synthesis with Argument Provenance for Multi-Turn Tool Use
Zichen Tian, Jinpeng Chen, Cheng Gong et al. (+2 more)
High-quality multi-turn tool-use data is essential for training agentic models, yet existing data synthesis methods often underrepresent the argument-level dependencies that are critical to long-horizon tool use. As a result, even when a model selects the correct tool, task execution may still fail because the model fills tool arguments with fabricated, stale, or weakly grounded values. To address this problem, we propose State-Guided Data Synthesis with Argument Provenance (SAP). SAP combines state guidance, tool-argument provenance constraints, and turn-level validation to efficiently construct tool-use trajectories with long-range dependencies and high accuracy. Using data generated by SAP, we build SAP-4B, which is highly competitive even when compared with much larger models across multiple
benchmarks. Source code, synthesized data, and trained weights are available at https://github.com/Zichen1024/SAP.
Read the full abstract ▾
All authors Zichen Tian, Jinpeng Chen, Cheng Gong, Suiyun Zhang, Rui Liu
Artificial Intelligence · Cross-listed from cs.NE
Programmable Cellular Automata
Ahmed Khalifa, Muhammad Umair Nasir, Matthew Siper et al. (+2 more)
Cellular automata is a local computation paradigm where complex behavior can arise from local interactions between simple functions. This paradigm has been used to explain many systems such as biological processes, traffic simulation, computer networks, etc. In games, cellular automata have been used in games such as SimCity and for the generation of spatial content such as caves or dungeons. However, creating effective local rules is hard and unintuitive. Cellular automata can be effectively evolved, but may still be hard to interpret. In this work, we introduce the concept of programmable cellular automata, where we represent the system as Python code. We also modularize the cellular automata into local functions and a decision function. Local functions take a local neighborhood and return a value, while the decision function takes the output of the local functions and decides the value of the next state. Separating the cellular automata into modules written in Python helps with understanding how these systems are working. We also explore adding global functions where they take the whole state and compute a function from it. We tested generating levels for three different games from the PCG
Benchmark. The results showed that global functions decrease the number of iterations that cellular automata need to solve a problem, and that we cannot find solutions for some problems with purely local functions. Looking into the generated functions, we can see common functions that have been used in different experiments, which not only helps us understand the generator but also helps us understand these games better and what is important for them.
Read the full abstract ▾
All authors Ahmed Khalifa, Muhammad Umair Nasir, Matthew Siper, Steve James, Julian Togelius
Artificial Intelligence · Cross-listed from cs.LG
VERPO: Verified Evidence Regularized Policy Optimization
Haijiang Li, Chengyu Lv, Yi Zhang et al. (+8 more)
Verifiable rewards improve
language models through reliable task-level feedback, but methods based on Group Relative Policy Optimization (GRPO) apply a sequence-level advantage uniformly across all
tokens. This coarse credit assignment reinforces or penalizes entire responses without identifying which local decisions to preserve, reinforce, or revise. Conversely, evidence-conditioned self-
distillation provides denser
token-level supervision, yet teacher imitation can transfer stylistic artifacts and miscalibrated confidence that destabilize training when misaligned with task success. We introduce VERPO, which converts evidence-conditioned guidance into reward-aligned
token-level credit assignment while retaining the outcome objective. VERPO decomposes teacher guidance into an evidence-free reference term and signed, evidence-induced corrections at each
token. A stopped controller combines selective acceptance,
token-wise localization, and cost-aware scaling by balancing
alignment with the local GRPO update direction against Fisher movement cost. Furthermore, we introduce Fisher Evidence Contrast (FEC), which attenuates nuisance shifts along an estimated evidence-presence direction through a regularized projection. Across five scientific
reasoning and tool-use tasks, VERPO prevents optimization collapse and consistently achieves the highest multi-task average across model backbones, yielding marked improvements particularly on smaller models over strong baselines. Qualitative diagnostics confirm that
token acceptance selectively targets
reasoning bottlenecks consistent with local reward
alignment and Fisher movement cost.
Read the full abstract ▾
All authors Haijiang Li, Chengyu Lv, Yi Zhang, Rui Qian, Zhibing Zhang, Xiangqing Shen, Junjie Yang, Yuchen Zhang, Wenyuan Jiang, Hanqing Hu, Cangqi Zhou
Artificial Intelligence · Cross-listed from cs.HC
Flawed but Memorable: Student Critical Reception of Interest-Personalized GenAI Analogies in Computing Education
Seth Bernstein and Naaz Sibia
Motivation: Undergraduate computing students increasingly turn to generative AI (GenAI) tools to understand abstract concepts through analogies. Analogies compare an unfamiliar concept to something familiar, but judging whether the comparison holds requires knowledge of both. GenAI may also embed assumptions about who the learner is. GenAI education research centers on output correctness, leaving students' critical reception of analogies largely unexamined. Method: We investigate how students evaluate the accuracy, appropriateness, and assumptions in GenAI-generated analogies, and their perceptions of interest-personalized versus generic technical explanations. Ten students with CS2 experience participated in a pre-survey, a think-aloud task with linked-list and recursion explanations, and a semi-structured interview grounded in the Paul-Elder framework. They judged accuracy, clarity, engagement, and trust separately. Results: Most participants described interest-personalized analogies as more engaging or memorable than generic technical explanations, while trust was mixed. Some trusted the tailored analogies more; others scrutinized them more closely or distrusted the tailoring. Participants with deep source-domain knowledge identified structural flaws requiring that knowledge to recognize. Because personalization and explanation format differed together, these findings do not isolate an effect of personalization alone. Implications: A familiar source flips the student's role. On the concept they are still learners, but on the familiar source they are the expert, and that is the position from which an analogy can be judged. We call this two-sided analogy
auditing. GenAI systems should ask what students know, not just what interests them, and treat a flawed analogy as something to inspect and fix rather than accept.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from q-fin.TR
Explainable Deep Learning for Price-Trade Dynamics: From Black-Box Forecasts to Effective Parametric Models
Manuel Naviglio and Fabrizio Lillo
Understanding the joint dynamics of prices and trades is central to
market microstructure, where returns and order flow interact through nonlinear and state-dependent mechanisms. Linear models are interpretable but may miss these effects, while deep neural networks improve
forecasting at the cost of transparency. We use neural networks as tools for structural discovery rather than only for prediction. A deep feed-forward network is trained on high-frequency returns and signed volumes for large- and small-tick stocks and compared with a linear VAR
benchmark. The neural network improves predictive performance, especially for returns, revealing nonlinear dependencies beyond the linear specification. Using Shapley-based explainability, we show that the dominant contributions are concentrated at the most recent lags. Model-implied responses are consistent with conditional averages reconstructed from the data. Unlike empirical averages, however, the neural-network decomposition isolates individual regressor contributions to the aggregate dependence. Lagged signed volume generates sign-preserving and saturating effects, consistent with nonlinear price impact and order-flow persistence. Lagged returns act as state variables: when the previous trade does not move the price, the model predicts continuation in the direction of past order flow, whereas non-zero returns generate attenuation or reversal. Building on these findings, we introduce a parsimonious SHAP-inspired nonlinear parametric model. It reproduces the main return-volume dependencies, outperforms the linear VAR
benchmark, and achieves performance comparable to the neural network. A multi-lag extension captures residual longer-
memory effects while preserving interpretability. Overall, explainability offers a route from black-box prediction to economically meaningful parametric models of price and trade dynamics.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
PhenoBench: Mapping What a Deeply Phenotyped Human Cohort Can Tell Us
Gal Sapir, Alon Diament, Adva Wolf et al. (+7 more)
Deeply phenotyped cohorts combine clinical, imaging, molecular, and wearable observations across timescales from seconds to years, but heterogeneous analyses are not directly comparable. We present PhenoBench, an executable
benchmark that turns deep-phenotyping measurements into explicit questions and controlled comparisons of information sources and predictive models. It is built around the Human Phenotype Project, with more than 13,000 participants at the initial visit. Each question fixes the target, population, timing, and allowed information; its evaluation contract specifies the split, metric, baseline, and claim boundary. PhenoBench defines 90 clinically grounded tasks across 15 domains and 26 input modalities. Across 160 matched regression comparisons spanning 52 tasks, six pretrained tabular models ranked above the evaluated task-specific baselines, including XGBoost and CatBoost, under a fixed single-
estimator protocol with bounded tuning. Giving each task equal weight, their mean advantage over ridge was 0.0103 $R^2$ (95% task-bootstrap interval, 0.0071-0.0136). We also evaluated 14
language models, collectively covering 40 tasks spanning phenotype recovery, classification, follow-up
forecasting, and participant ordering. Without cohort-specific fitting,
language models made informative predictions on some tasks but showed task-specific capability gaps, shared failures of scale, and rarely surpassed task-specific ridge or logistic regression models fitted on the same input fields. PhenoBench provides a versioned, auditable evaluation system where new questions, measurements, and models can be added without redefining existing comparisons.
Read the full abstract ▾
All authors Gal Sapir, Alon Diament, Adva Wolf, Doron Yaya-Stupp, Dikla Gelbard Solodkin, Dana Azouri, Anat Etzion-Fuchs, Guy Lutsker, Eran Segal, Hagai Rossman
Artificial Intelligence
LayerRoute: Action-Conditioned Mixture-of-Layers Routing for Vision-Language-Action Policies
Zheng Lu, Haoran Liao, Wanqi Zhong et al. (+9 more)
Vision-
Language-Action (VLA) policies leverage pretrained
vision-
language models (VLMs) to guide action generation for
robot control. VLMs provide hierarchical visual-semantic representations that evolve across layers, from local visual geometry to abstract,
language-aligned semantics; different manipulation tasks may therefore require different mixtures of layer representations. Meanwhile, the action module maintains intermediate representations that evolve throughout action computation and may provide useful information for subsequent decisions. However, existing VLA interfaces offer limited flexibility in representation access: VLM information is exposed through fixed layer assignments for each action layer, while intermediate action states are only propagated implicitly through residual streams without explicit reuse. We introduce LayerRoute, an action-conditioned representation routing interface that enables adaptive access to VLM layers and action representations. The Layer Mixture Router dynamically forms mixtures of cached VLM representations, while Action-State Reread reuses earlier action representations. Across diverse simulation and real-world
benchmarks, LayerRoute consistently improves StarVLA-$\pi$ and $\pi_{0.5}$, achieving up to 7.2 gains on LIBERO Long with only 0.31% / 3.87% additional parameters. Ablation studies validate the benefit of action-conditioned layer routing, while routing analyses reveal structured allocation patterns across action layers and task settings.
Read the full abstract ▾
All authors Zheng Lu, Haoran Liao, Wanqi Zhong, Yunhe Ni, Lijie Wang, Xingjie Fan, Zhisheng Chen, Yantang Qu, Meijia Chen, Tianyu Xin, Zirui Song, Yiming Li
Artificial Intelligence · Cross-listed from cs.LG
FedSubMuon: Communication-Efficient Federated LLM Fine-Tuning via Structured Subspace Muon
Shaolong Chen, Youming Tao, Shuzhen Chen et al. (+3 more)
Federated fine-tuning adapts large
language models (LLMs) to decentralized client data, but its scalability in cross-device training is often limited by the high communication cost.
Muon is an optimizer that improves optimization performance by orthogonalizing momentum for
matrix-valued parameters. Existing
federated Muon methods demonstrate the benefit of
matrix-aware optimization in
federated learning, but still require transmitting full layer-size updates and optimizer state. A natural way to reduce communication is to directly apply
Muon to LoRA factors, but this changes the optimized object and weakens
Muon's
matrix-aware update geometry. We propose FedSubMuon, a communication-efficient
federated Muon fine-tuning method that optimizes compact coefficient
matrices within shared structured subspaces. This design keeps
Muon on a single
matrix-valued trainable object, while reducing the client upload to compact coefficient
matrices. We further introduce FedSubMuon-GT, an accuracy-oriented extension that uses projected gradients to adapt tracked subspace bases toward task-relevant gradient directions. Experiments on instruction tuning and mathematical
reasoning show that FedSubMuon-GT achieves the best overall accuracy on four of five
dataset-model pairs, while FedSubMuon performs best under all matched communication budgets. On Dolly-15K, the closest communication baseline requires 5.5 times and 1.4 times more total communication on Llama-1B and Qwen-4B, respectively.
Read the full abstract ▾
All authors Shaolong Chen, Youming Tao, Shuzhen Chen, Falko Dressler, Qingqing Ye, Di Wang
Artificial Intelligence · Cross-listed from cs.LG
ACE: Adapter Consolidation across Experts for Parameter-Efficient Fine-Tuning of MoE LLMs
Ahin Lee, Sehyun Yun, Joonha Park et al. (+1 more)
Parameter-efficient fine-tuning (PEFT) of mixture-of-experts (MoE) models commonly attaches a separate low-rank adapter to each expert. This expert-wise design fragments adaptation in three ways: capacity is split across narrow low-rank updates, gradient supervision becomes sparse and imbalanced under sparse routing, and execution is decomposed into many small GEMMs. We find that such expert-wise separation is often unnecessary, as subsets of LoRA adapters become functionally similar during fine-tuning, revealing redundancy among expert-specific adapters. Based on this redundancy, we propose ACE (Adapter Consolidation across Experts), which groups redundant experts and replaces their expert-specific adapters with group-shared higher-rank LoRA modules under the same PEFT budget. ACE further introduces grouped adapter execution, which consolidates fragmented expert-wise adapter computations into fewer, larger group-level GEMMs. Across evaluations covering 12
datasets and four MoE backbones, ACE achieves the highest observed mean accuracy among the parameter-matched PEFT methods on the three backbones with complete baseline coverage, while providing $1.31\times$ to $1.48\times$ wall-clock training speedup over expert-wise LoRA without increasing peak
memory. Our code is available at https://github.com/UbiquitousAILab/ACE.
Read the full abstract ▾
All authors Ahin Lee, Sehyun Yun, Joonha Park, Taesik Gong
Artificial Intelligence
Generating Instance Generators in PDDL Planning
Nicola J. Müller, Naya Rudolph, Katharina Stein et al. (+3 more)
PDDL, the de-facto standard
language in the AI Planning community, is designed to specify planning domains: sets of instances that share the same predicates and action schemas. Yet it does not provide any means to specify the actual instance set, i.e., legality constraints on initial states and goal conditions, as well as possibly domain subset constraints specifying an instance subset we are interested in. One consequence of this is that instance generation has always been ad-hoc, with manually written domain- and subset-specific instance generators. Recent work has started to address this, through
reasoning and learning methods that however suffer from scalability limitations. Here we introduce an alternative approach, leveraging LLMs to generate instance-generation programs, with built-in soundness guarantees through prescribed checks. We show that these automatically generated instance generators return large numbers of sound and diverse instances efficiently.
Read the full abstract ▾
All authors Nicola J. Müller, Naya Rudolph, Katharina Stein, Jörg Hoffmann, Ayal Taitler, Timo P. Gros
Artificial Intelligence · Cross-listed from math.OC
The Role of Gradient Modification in Heavy-Tailed Nonconvex Stochastic Min-Max Optimization
Tianxi Zhu, Yi Xu and Xiangyang Ji
Stochastic min-max optimization has attracted increasing
attention due to its applications in modern machine learning, while existing theoretical studies mainly rely on the bounded variance assumption for
stochastic gradients. Under heavy-tailed noise, where
stochastic gradients only possess a finite $p$-th moment for $p\in(1,2]$, gradient clipping or normalization is commonly believed to be necessary to guarantee convergence. In this work, we revisit
stochastic min-max optimization under heavy-tailed noise and provide a comprehensive theoretical study of
stochastic gradient descent ascent (SGDA). We first show that vanilla SGDA, without any modification to its update rule, can converge under heavy-tailed noise in both nonconvex-strongly-concave (NC-SC) and nonconvex-concave (NC-C) settings, establishing the first convergence guarantees for SGDA in these regimes. Beyond unregularized problems, we further investigate regularized
stochastic min-max optimization, where directly incorporating gradient normalization into proximal updates is nontrivial due to the incompatibility between normalization and proximal structures. We overcome this difficulty by developing new clipping-free
algorithms, i.e., Stoc-TRGDAM and Stoc-TRGDmax, and they both can achieve the optimal dependence on the target accuracy without using gradient clipping.
Read the full abstract ▾
Artificial Intelligence
Explaining AI Agents Through Execution Traces
Vittoria Vineis, Fabiano Veglianti, Lorenzo Antonelli et al. (+3 more)
AI
Agents are increasingly deployed in real-world settings, where they interact with external tools and make sequential decisions with limited human oversight. This creates a pressing need for reliable and auditable explanations of what an
agent did and why. However, traditional Explainable AI (XAI) methods fall short of providing the process-level transparency required for such interactive, multi-step systems, motivating a paradigm shift toward approaches specifically designed for AI
Agents. To address this gap, we present a post-hoc XAI framework that transforms a lengthy
agent's execution trace into a structured report and a faithful natural-
language explanation explicitly grounded in its observable behavior. Because it relies solely on execution traces, the framework applies across different
agent architectures, environments, and tasks. Human and automated evaluations across multiple
benchmarks and architectures show that our framework produces high-quality, trace-faithful explanations while reliably identifying unsupported claims, unjustified actions, and evidence gaps, outperforming naive LLM-generated explanations.
Read the full abstract ▾
All authors Vittoria Vineis, Fabiano Veglianti, Lorenzo Antonelli, Claudia Di Carlo, Matteo Silvestri, Gabriele Tolomei