Artificial Intelligence
UniMoMo: Expert Merging-Based MoE Acceleration for Large Recommendation Models
Lei Xin, Bin Gu, Peize Li et al. (+8 more)
Sparse mixture-of-experts (MoE) layers expand recommendation capacity through conditional computation, yet a trained checkpoint still stores and routes over its full expert bank. We study a deployment problem: convert that checkpoint to a smaller standard MoE under an explicit expert budget, without adding a compression-specific online module. To address this, we introduce UniMoMo, a post-training compression framework formulated as a constrained graph coarsening problem. Rather than relying on parameter distance, UniMoMo groups experts based on their functional similarity, using an unlabeled calibration set to measure how similarly experts respond to shared recommendation states. To prevent performance degradation, we introduce a layer-adaptive protection mechanism that restricts the merging of high-traffic experts based on their routing exposure. Across Amazon Beauty, KuaiRec, and TenRec with 2, 4, and 6 MoE blocks, the final four-expert checkpoints obtain source-relative five-run mean NDCG@10 ratios of 99.92%–102.30% and measured A100 speedups of 1.28$\times$–1.63$\times$. An aggressive two-expert, top-1 operating point obtains ratios of 98.36%–104.24% and speedups of 1.47$\times$–2.21$\times$. These endpoint results evaluate the complete conversion-and-adaptation workflow and show that a trained recommendation MoE can be exported at multiple serving budgets.
Read the full abstract ▾
All authors Lei Xin, Bin Gu, Peize Li, Zitong Wang, Jianbo Zhao, Changjiang Jiang, Chao Huang, Xuyang Zhao, Zunhai Su, Fanhu Zeng, Zhenglun Kong
Artificial Intelligence
MedCalc-R1: Knowledge-Guided Reward Framework for Medical Mathematical Reasoning
Haotian Wang, Lian Yan, Xingzhi Yao et al. (+4 more)
In Reinforcement Learning with Verifiable Rewards (RLVR) frameworks for mathematical
reasoning tasks, floating-point results are typically evaluated using a tolerance-based reward. However, this strategy suffers from challenges such as difficulty in threshold calibration, unstable training dynamics, and limited accuracy, especially in clinical scenarios. To address these limitations, we propose a knowledge-guided hybrid reward framework (MedCalc-R1). Specifically, we introduce a knowledge verification reward mechanism that enforces explicit generation of computational formulas, which are further validated by an external verifier to enhance interpretability and
reasoning reliability. Furthermore, we design a hybrid soft-hard reward scheme combining a hard constraint based on clinical safety thresholds with a soft, precision-sensitive reward that progressively guides learning within the acceptable range. Experimental results demonstrate that our method significantly outperforms existing baselines in both
reasoning accuracy and generalization capability, validating the effectiveness and applicability in safety-critical domains.
Read the full abstract ▾
All authors Haotian Wang, Lian Yan, Xingzhi Yao, Fanshu Meng, Ye He, Jingchi Jiang, Yi Guan
Artificial Intelligence
Business Arena: Benchmarking LLM Agents in a Realistic Marketplace
Yijun Pan, Yukun Lian, Kunyu Shi et al. (+5 more)
Running a business is a challenging form of intelligent work. Operators must infer opportunities from partial signals, commit capital under uncertainty, adapt to delayed outcomes in a changing
market, and satisfy regulatory obligations before trading legally. Frontier LLM
agents can increasingly complete complex workflows, yet business-related capabilities are rarely evaluated in existing
agent benchmarks. We introduce Business Arena, a controlled environment where an AI
agent runs a cross-border shop, buying from suppliers and selling to buyers over a long horizon. We ground the arena in real Alibaba.com sourcing data and
market conditions calibrated from authoritative sources. Delayed and coupled consequences make individual business decisions difficult to judge, but their combined outcome is measurable through profit. Because profit alone cannot explain why an
agent succeeds or fails, we compare
agents with human-designed strategies to estimate available opportunity, use skill-level metrics to reveal underlying strengths and weaknesses, and trace realized gains and losses to the actions that produced them. We use mechanism ablations to establish that strong results reflect genuine business intelligence rather than neglect or simulator-specific shortcuts. We evaluate 15 frontier models and find a ninefold difference in mean final net worth. Even the best model falls behind human-designed strategies, indicating that business operation remains challenging for LLM
agents. Skill-level analysis reveals operating styles, from margin-focused premium sellers to high-turnover wholesalers and customer-service specialists, while action-level attribution identifies the sourcing, pricing, and recovery decisions that create or destroy value. Together, Business Arena takes a first step toward a realistic and trustworthy testbed for evaluating end-to-end business
agents.
Read the full abstract ▾
All authors Yijun Pan, Yukun Lian, Kunyu Shi, Junbo Li, Hongwei Xue, Sicong Xie, Guannan Zhang, Xiaoying Xing
Artificial Intelligence · Cross-listed from cs.RO
RAG-Based Auto-Configuration for Industrial Fieldbus Devices
Aadil Gani Ganie, Saad Ezzini and Naveed Farooz Marazi
Industrial device commissioning requires engineers to manually extract hundreds of protocol-specific parameters from heterogeneous PDF manuals and transcribe them into supervisory control systems, a time-intensive, error-prone workflow. This paper presents SysName, a production-oriented pipeline that automates device configuration end-to-end for Modbus RTU, OPC-UA, Profibus DP, and CANopen. It builds a hybrid dense-sparse
retrieval index augmented by an ontology graph derived from ECLASS, AAS, and SOSA/SSN, using a BGE-M3 encoder with a cross-encoder reranker to surface relevant manual passages. A local LLM (T=0.1) generates ontology-aligned JSON-LD configurations via protocol-specific
prompts and a four-step repair pipeline. A two-stage abstention gate, combining a reranker-score threshold and an IRI resolution ratio, blocks unsafe LLM invocations and filters low-coverage configurations before SHACL validation. On a gold set of 28 field-level queries, the hybrid retriever reaches 0.96 HitRate@10, and the reranker raises MRR@10 from 0.56 to 0.63 with perfect score separation for abstention. The generator attains field-level F1=0.87 with exact match on 9 of 12 runs. End-to-end runs on an H100 GPU complete in 2.6-6.6s per device with zero unsafe writes and zero silent failures on a five-device
benchmark; every unsuccessful run is flagged by abstention or deployment verification. Component-wise evaluation localises the single systematic failure to OPC-UA generation, invisible to end-to-end metrics alone. A case study commissions a physics-simulated Universal
Robots UR5e
robot from unmodified vendor documentation (254-page manual, 8-page register list, 496 chunks), reaching field-level F1=1.0 over three runs with read-back and joint-consistency verification. An ablation study and comparison with five industrial-LLM systems complete the analysis.
Read the full abstract ▾
Artificial Intelligence
Walking through Discussions: A Mobile Visual Analytics System for In-Situ Group Discussion Analysis
Yiping Sun, Ziyao Kang, Wei Zeng et al. (+2 more)
Group discussion-based teaching is widely used to foster collaborative learning, yet teachers in physical classrooms often struggle to simultaneously monitor multiple groups and quickly diagnose a target group before intervening. Existing visual analytics tools primarily support post-hoc analysis on desktop, providing limited support for in-situ walk-around teaching. To address this gap, we present MobileGroupVis, a mobile visual analytics system for in-situ analysis of classroom group discussions. MobileGroupVis integrates multi-group monitoring, single-group diagnosis, and instructional intervention into a concise analytical workflow tailored for small-screen touch interaction. The system is powered by a lightweight streaming analysis pipeline that converts group audio into structured discussion data and further extracts interaction patterns, topic progression, and topic deviation through a dialogue analysis module. To enable both glanceable overview and traceable diagnosis, we design six coordinated views, including a compact glyph that visually encodes word count, interaction intensity, and topic deviation for efficient cross-group comparison and anomaly localization, along with detailed views for opinion
evolution, interaction dynamics, topic coverage, and dialogue records. We evaluate MobileGroupVis through two case studies and expert interviews. The results provide preliminary evidence that MobileGroupVis supports teachers in understanding discussion processes, identifying groups in need of
attention, and facilitating in-class intervention.
Read the full abstract ▾
All authors Yiping Sun, Ziyao Kang, Wei Zeng, Minli Wu, Jiazhi Xia
Artificial Intelligence · Cross-listed from cs.CV
REVEAL: A Rubric-Guided Agent for Explicit Evidence Sufficiency Verificationin Long-Video Question Answering
Caijun Yan, Yang Zhou, Meixing Shi et al. (+4 more)
Recently,
retrieval-augmented and
memory-augmented methods have emerged as two promising paradigms for long-
video question answering. However, existing methods typically rely on rigid, fixed-length temporal chunking (e.g., 10s) and static offline
memory banks, which not only fragment coherent continuous events but also fail to adapt during real-time
reasoning. Moreover, whether using multi-scale summaries or
multimodal knowledge graphs, current approaches prioritize
retrieval relevance while overlooking evidence sufficiency, often stopping to answer once only semantically relevant clues are retrieved, even when key temporal, causal, or fine-grained action evidence is still missing. To tackle these challenges, we propose REVEAL, a rubric-guided
agent framework. As a foundation, we introduce an adaptive visual-similarity-based preprocessing pipeline that groups visually coherent adjacent frames into natural event units to construct an offline-online
video memory—capturing global
video context offline while dynamically maintaining question-conditioned
memory online. Built upon this structured
memory, REVEAL uses an automatically constructed rubric library to explicitly verify whether retrieved evidence satisfies sufficiency criteria, pinpoints missing clues upon verification failure, and directs targeted re-
retrieval for complementary information. Without any extra training, REVEAL consistently outperforms both closed-source and open-source state-of-the-art methods across extensive experiments. These results show that explicitly verifying evidence sufficiency, rather than stopping at semantic relevance, retrieves the decisive clues that prior methods miss and yields more reliable long-
video reasoning.
Read the full abstract ▾
All authors Caijun Yan, Yang Zhou, Meixing Shi, Haoran Sun, Yichen Li, Yuxiang Cai, Yankai Jiang
Artificial Intelligence · Cross-listed from cs.CL
Mitigating Gender Bias in English to Romanian Machine Translation
Ioana Grigore and Sergiu Nisioi
Machine translation (MT) systems often fail to correctly translate gender, especially when converting from a gender-neutral
language like English to a gendered target
language such as Romanian. This bias results in translations that default to masculine forms or reinforce gender stereotypes. We propose a hybrid pipeline to mitigate this issue by combining large
language model (LLM)-based gender classification with neural machine translation (NMT). Our system uses a fine-tuned LLM to detect the intended gender of target words in English sentences and insert inline gender hint tags. These tagged sentences are then passed to a
Transformer model fine-tuned to generate morphologically correct Romanian translations. To support this, we introduce three novel
datasets for gender disambiguation and translation. Our approach improves gender accuracy on the WinoMT and WinoGender
benchmarks by over 40 percentage points compared to a baseline MT system. This is the first method to explicitly address and evaluate gender bias in English-Romanian MT using both LLM inference and tag-aware translation.
Read the full abstract ▾
Artificial Intelligence
ForestBench: A Unified Graph Framework for Evaluating Multi-Agent Collaboration
Guo Chen, Ziwen Li, Reed Li et al. (+4 more)
Multi-
agent systems (MAS) built on Large
Language Models (LLMs) are proliferating rapidly, but their heterogeneous execution traces provide no common basis for evaluation across methods. Outcome-only
benchmarks discard collaborations, whereas LLM-as-Judge evaluation requires additional, model-dependent inference and can vary with the LLM and rubric. We introduce a generalizable evaluation framework that maps native MAS traces into a shared space of unified collaboration graphs, enabling different methods to be evaluated under the same representation, reference set, and metric panel. Candidate graphs are compared with a query-specific reference forest. Each forest is a
benchmark-provided collection of verified-success graphs: it records diverse ways in which representative MAS methods can complete the task, rather than prescribing a unique optimal process. Instantiating the framework as ForestBench, we filter $844$ collaboration-necessary queries from seven public
datasets, precompute ten successful target-conditioned reference graphs per query, and evaluate six representative MAS frameworks. Controlled backbone, reference-construction, and perturbation studies test the stability and scope of evaluation. Once the
benchmark forests are built, ForestBench scores a trace in milliseconds without further LLM inference, providing a reusable structural basis for comparing diverse MAS collaboration traces.
Read the full abstract ▾
All authors Guo Chen, Ziwen Li, Reed Li, Yu Lu, Haibo Shi, Bingbing Xu, Junjie Huang
Artificial Intelligence
Unaccountable Delegation, Fading Skills: Mapping the Risks of Workplace AI Agents
Gabriele La Malfa, Lakmal Meegahapola, Edyta Bogucka et al. (+4 more)
To anticipate socio-technical risks from AI
agents, organizations need taxonomies to classify them. However, existing AI risk taxonomies focus on broad risks and do not capture job-specific risks introduced by
agents. To address this gap, we make three main contributions. First, we developed a multi-layer framework from a literature review of AI
agents. The framework models three core components and their interactions:
agents, goals, and environment. Second, we embedded this framework in a structured
prompt and applied it to descriptions of 2,078 job tasks from the O*NET database, producing 8,356 risk scenarios labeled by severity and deployment mode (automation or augmentation). We validated these scenarios with 45 workers across 10 job roles and an independent LLM judge, confirming their plausibility and
alignment with job tasks. Finally, we extended an existing taxonomy to create a 15-category taxonomy of workplace AI
agent risks that covers all our risk scenarios. Our analysis highlights four findings. First, augmentation is not inherently safe because overreliance on
agents can gradually erode workers' skills and oversight. Second, Erroneous
Agent Actions accounts for the largest share of risk scenarios and has the highest concentration of severe risks. Many arise at the human-
agent boundary. Third, automation is associated mainly with organizational risks, while augmentation is associated mainly with risks to workers. Fourth, workers found our taxonomy easier to use for a risk classification task than two other taxonomies and preferred it in 64% of non-tied comparisons with a recent generative AI risk taxonomy. These findings show that workplace AI
agent risks do not arise from
agents alone; they also depend on how people work with
agents and how
agents are deployed. Safer workplaces require not only safer
agents but also carefully designed human-AI
agent collaboration.
Read the full abstract ▾
All authors Gabriele La Malfa, Lakmal Meegahapola, Edyta Bogucka, Jie M. Zhang, Michael Luck, Elizabeth Black, Daniele Quercia
Artificial Intelligence · Cross-listed from cs.CV
Population-Scalable Multi-Agent World Modeling
Renjie Zhao, Yuxiang Wu, Mingyu Zhang et al. (+9 more)
World models have recently achieved impressive progress in visual prediction and interactive generation, but extending them to multi-
agent environments introduces a fundamental scalability challenge. Existing methods generally assume a fixed number of
agents during training and inference, which ties the model to a pre-determined
agent population and limits inference-time scalability. Our key insight is that cross-view consistency should arise from a shared world state whose
evolution does not assume a predefined number of
agents, while
agent-specific observations should be generated by querying this state through a unified rendering interface. Based on this insight, we propose Khora, a scalable multi-
agent world model that supports inference-time expansion to arbitrary numbers of
agents without retraining. Our framework decouples world-state
evolution from visual rendering and introduces a population-agnostic rendering mechanism for incorporating other
agent information. This design maintains cross-view consistency through the shared world state rather than through dense interactions among observation streams inside the expensive
video generator, enabling approximately linear practical scaling with the number of queried views. Qualitative experiments demonstrate that our approach generalizes to unseen numbers of
agents while maintaining visual quality and multi-
agent consistency. We further implement a real-time interactive system to demonstrate scalable open-world simulation.
Read the full abstract ▾
All authors Renjie Zhao, Yuxiang Wu, Mingyu Zhang, Jiaxin Li, Sisi Li, He Li, Yimin Sheng, Tianxi Tan, Zhenkai Zhang, Jiao Liang, Jianyi Zhu, Yong-Lu Li
Artificial Intelligence
SDDBMs: Soft Denoising Diffusion Bridge Models
Shiyi Qi, Kun He and Mingmou Liu
Diffusion bridge models leverage Doob's \(h\)-transform to construct
stochastic transports between arbitrary endpoint distributions, and have shown strong potential in image-to-image translation and restoration. However, most existing bridge models rely on hard endpoint conditioning, which forces the terminal state to match a prescribed target exactly. This hard constraint induces terminal-boundary singularities: the terminal law collapses to a Dirac measure, and the resulting drift coefficients become ill-conditioned near the endpoint. In this paper, we propose Soft Denoising
Diffusion Bridge Models (SDDBMs), a generalized framework that regularizes
diffusion bridges directly at the level of their terminal constraints. Instead of imposing an exact endpoint, SDDBMs prescribe a non-degenerate Gaussian terminal marginal under the transformed path measure, with a flexible terminal center and variance. Starting from this prescribed marginal, we develop a complete closed-form construction of the soft bridge, including the Gaussian terminal reweighting and soft \(h\)-function, the induced Gaussian forward marginals and \(\mathbf{x}_0\)-free dynamics. Theoretically, SDDBMs provide a unified probabilistic perspective that encompasses existing
diffusion bridge models, including DDBMs, GOUB, and UniDB, as special cases under specific parameter choices. Extensive experiments on image restoration tasks demonstrate that SDDBMs achieve improved numerical stability and superior generation quality over existing bridge-based methods.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
CDGC-Net: 3D Medical Image Segmentation with Cooperative Dual-Scale Self-Attention and Grouped Channel Modeling
Zheyang Jing, Qin Lu, Jianwang Li et al. (+3 more)
Accurate 3D medical image
segmentation requires the integration of long-range anatomical context with fine boundary detail. Existing methods often model global and local features in separate modules or feature levels and perform channel recalibration independently. This may cause semantic mismatch between global context and local boundaries, insufficient channel relationship modeling, weak spatial-channel interaction, and redundant representations. We propose CDGC-Net, a 3D medical image
segmentation network that combines cooperative dual-scale spatial
attention with grouped hierarchical channel modeling. With-in each CDGC block, Cooperative Dual-Scale Self-
Attention (CDSA) assigns
attention heads to parallel local-window and global-sparse branches. The two branches capture fine spatial details and long-range anatomical context at the same feature level. Their outputs are concatenated into an $N\times C$ spatial representation and directly passed to Grouped Hierarchical Channel
Attention (GHCA). GHCA organizes the channels into $r$ groups and models both within-group and cross-group dependencies. CDSA and GHCA reuse a shared key projection to maintain a consistent feature reference. Residual feature
alignment subsequently integrates the refined features with the original representation. On the Synapse, ACDC, BraTS, and LA
datasets, CDGC-Net achieved mean DSC values of 86.96%, 92.91%, 82.56%, and 93.52%, respectively, exceeding the next-highest reported values by 0.39, 0.47, 0.17, and 0.32 percentage points. CDGC-Net contains 25.83M parameters and 28.62G FLOPs for an input size of $64\times128\times128$, reducing these quantities by 39.87% and 40.30%, respectively, relative to UNETR++. These results indicate a favorable trade-off between
segmentation accuracy and computational complexity.
Read the full abstract ▾
All authors Zheyang Jing, Qin Lu, Jianwang Li, Yujie Yang, Chen Yi, Shaofeng Jiang
Artificial Intelligence
FailForge: Distilling Procedural Competence from Persistent Failures into Code Agents
Dongyi Lv, Fushun E, Aichen Cai et al. (+8 more)
Rejection sampling fine-tuning (RFT) is widely used to train code
agents by generating trajectories on verifiable software engineering tasks, retaining those that pass the tests, and fine-tuning on the successful rollouts. However, even strong code
agents repeatedly fail on a substantial fraction of such tasks, and standard RFT simply discards these failures. The discarded samples are precisely the hardest and most informative ones, drawn from verifiable instances that are costly to curate. Stronger base models may reduce the number of failures, but the remaining hard cases still define the frontier for further improvement. We propose FailForge, an agentic framework that converts failed rollouts into training signal. For each failed instance, an
agent diagnoses the failure from error feedback and execution traces, distills the diagnosis into a concise and actionable skill, and injects the skill into the
agent context for a guided second attempt. Trajectories that succeed under skill guidance are folded back into the RFT corpus. Crucially, the skill is removed at training time, so the model internalizes the recovered behavior rather than relying on external hints at inference. FailForge recovers over 26% of previously failed instances at marginal additional cost, and training Qwen3.5-4B on the augmented corpus improves the SWE-bench Verified resolve rate by 6.6 points over a strong RFT baseline, with gains concentrated on the hardest problems.
Read the full abstract ▾
All authors Dongyi Lv, Fushun E, Aichen Cai, Liang Huang, Ya Zhang, Qiuyu Ding, Canhui Wu, Zhi Wang, Yuesong Zhang, Jiaqi Wang, Nan Duan
Artificial Intelligence
VoxZip: Semantic-Anchored Temporal KV Cache Compression for Long-Context Audio Inference
Wenxu Jia, Dongjie Fu, Xize Cheng et al. (+6 more)
Recent advancements in
Speech Large
Language Models have demonstrated remarkable capabilities in understanding complex audio tasks. Despite this progress, their long-context inference remains severely bottlenecked by prohibitive KV cache
memory demands. Existing text-centric compression methods struggle here, often disrupting
speech continuity or discarding crucial semantic cues. To address this, we propose VoxZip, a train-free, two-stage semantic-anchored KV cache compression framework. The first stage uses automatic
speech recognition (ASR) transcriptions as explicit semantic anchors to temporally align, compress, and fuse audio
tokens, significantly reducing the initial KV cache while elevating
token information density. To further improve the compression ratio, the second stage employs a dynamic filtering strategy based on temporally decayed accumulated
attention to evict non-essential
tokens while mitigating early-
token bias. Comprehensive evaluations on Qwen3-Omni across six diverse audio
benchmarks demonstrate the superiority of our approach. VoxZip excels in long-audio
reasoning and consistently maintains high-fidelity perception on short-form tasks. Notably, it sustains over 90% of the uncompressed baseline performance even under an aggressive 20x KV cache compression in long-context scenarios. Furthermore, at a 4x compression ratio, VoxZip yields a 1.9x increase in inference throughput alongside a 3.3x reduction in peak
memory overhead. Code and models will be available at https://github.com/MM-
Speech/VoxZip.
Read the full abstract ▾
All authors Wenxu Jia, Dongjie Fu, Xize Cheng, Fangming Feng, Linjun Li, Wenshi Chen, Yingming Li, Zhou Zhao, Tao Jin
Artificial Intelligence · Cross-listed from cs.CV
On-Device Multi-Species Malaria Detection with Uncertainty-Calibrated Slide-Level Aggregation
Idaya Seidu, Ahmed Tahiru Issah, Charles B. Delahunt et al. (+1 more)
Malaria remains a leading cause of mortality in resource-limited settings, where expert microscopists are scarce. Automated diagnosis based on microscopy images thus has strong potential to improve care delivery. But for an
algorithm to deploy, a necessary requirement is that it meet a suite of non-obvious (from a machine learning (ML) perspective) clinical constraints. Therefore, in close consultation with a national health center we developed a malaria diagnosis pipeline which addresses key requirements listed by the health care center but typically ignored in the ML malaria literature. In particular, it includes: (i) stopping criteria (to reduce image acquisition and time-to-result); (ii) human-in-the-loop functionality (for review and accountability); (iii) multi-species discrimination (since treatment varies by species); (iv) thick film detection (standard for microscopy); (v) computationally-efficient uncertainty calculations (to aid clinician review); and (vi) an edge device platform (since internet can be spotty in this catchment area). The mobile system performs all inference on-device using YOLOv13n deployed via TensorFlow Lite. It detects four species and white blood
cells from Giemsa-stained thick blood smear images, aggregating per-image detections into slide-level parasitemia with World Health Organization (WHO)-standard quantification. This paper highlights these various clinical constraints and offers methods to address them. Evaluated on 2,739 annotated images across all four species, the system achieves mAP@0.5 of 0.863, per-image parasite count correlation of r = 0.812, slide-level r = 0.951 (soft counting, 10 images/slide), and runs entirely offline with a pipeline time of 10.27 +- 1.65 s per image.
Read the full abstract ▾
All authors Idaya Seidu, Ahmed Tahiru Issah, Charles B. Delahunt, Carine Mukamakuza
Artificial Intelligence
Deep probabilistic logic programming for diagnostic reasoning from incomplete information: A case study in stroke detection
Felix Weitkämper, Monchito Avila, Elizabeth Nanjala et al. (+2 more)
In medical applications, raw data is frequently associated with significant
privacy concerns, lending particular importance to the encoding of summary statistics from the literature. On the other hand, deep learning has become an invaluable tool for assessing symptoms based on visual or auditory sensor data. DeepProbLog allows for an extensible neuro-symbolic approach that accommodates connectionist components to analyse patient images within a transparent and rigorous probabilistic framework, namely probabilistic logic programming under the distribution semantics. Framed as a case study in stroke detection from
multimodal data, this contribution explores the pathway from summary statistics available in the literature to a DeepProbLog-based diagnostic system. It suggests a workflow using established maximum
entropy techniques to complete available probabilistic information and the probabilistic logic programming system ProbLog 2 to move from the
entropy-maximising causal model to a discriminative neuro-symbolic model expressible within DeepProbLog. The relative performance of models derived from less complete data is analysed alongside the potential of the probabilistic inductive logic programming system ProbFOIL 2 for compressing large discriminative models, and the perspectives and implications of using DeepProbLog for diagnostic
reasoning are discussed.
Read the full abstract ▾
All authors Felix Weitkämper, Monchito Avila, Elizabeth Nanjala, Siska, Grace Zawadi
Artificial Intelligence
Discovering Diverse Planning Policies for Multimodal Embodied Agents with Quality-Diversity Optimization
Pengfei Xu, Yong Liu, Xiaoya Nan et al. (+2 more)
Multimodal embodied agents are increasingly required to solve long-horizon tasks by integrating visual observations, textual goals, and interaction history into closed-loop decision making. However, state-of-the-art large-model-based planners often rely on a single dominant planning style during execution. Once this execution mode becomes ineffective, the
agent may remain stalled for many steps, repeatedly interacting with the environment without making meaningful progress. We address this limitation by proposing a Quality-Diversity (QD) framework for discovering diverse planning policies for
multimodal embodied agents. The proposed method treats planning-policy templates as evolvable individuals and organizes them into a behavior-indexed archive rather than collapsing search to a single
prompt style. In the offline stage, rollout trajectories are summarized into structured success and failure experiences, which guide policy variation through recombination and experience-guided mutation. The resulting policies are mapped into a behavior space defined by interaction intensity and goal-directedness, and the highest-quality policy in each niche is retained in the archive. In the online stage, the
agent executes one policy at a time while monitoring task progress. When persistent stall is detected, the system rolls back to the latest checkpoint and switches to a behaviorally distinct archive policy to resume execution. Experiments on the ThreeDWorld transport
benchmark show that the proposed framework improves both task success and interaction efficiency over representative baseline planners. These results suggest that discovering diverse policy repertoires is an effective way to support adaptive
multimodal planning and online failure recovery.
Read the full abstract ▾
All authors Pengfei Xu, Yong Liu, Xiaoya Nan, Qiang Yang, Peilan Xu
Artificial Intelligence · Cross-listed from cs.CV
A Combined Feature-Based Framework for Disguise and Spoofing Detection in Face Recognition Systems
Sangiya Pararajasingham
Face recognition systems face two distinct, commonly-separated failure modes: spoofing, where an impostor presents a photograph or
video of an authorized user, and disguise, where a legitimate user is rejected because their appearance differs from their enrolled template due to accessories, facial hair, illumination, or pose. This paper proposes and compares five combined feature-extraction and classification pipelines that address both problems within a single framework: PM (PCA and Minimum Euclidean Distance, MED), LPM (Local Binary Patterns with PCA and MED), HPM (Histogram of Oriented Gradients with PCA and MED), SM (Speeded-Up Robust Features with MED), and HM (Harris corner features with MED). Each pipeline follows a common two-phase process comprising pre-processing, feature extraction, feature filtering, and classification. The methods were trained on 115 subjects drawn from the FEI, Disguised Faces Database, and NUAA databases and evaluated on six test conditions covering mixed appearances, frontal faces, dark illumination, left- and right-turned poses, and photo-spoofing attempts. The HOG-based pipeline (HPM) achieved the most consistent performance across conditions, with 94.59% accuracy on mixed-appearance disguise, 81.5-93.2% across pose and illumination variants, and 91.67% on spoofing, while the LBP-based pipeline (LPM) achieved the second-highest spoofing-detection accuracy (93.2%), behind PM (96.67%), but weaker robustness to pose change. These results reveal a measurable trade-off between spoof sensitivity and disguise robustness among classical feature representations, motivating the deep-learning and cross-database extensions discussed in the concluding sections.
Read the full abstract ▾
Artificial Intelligence
Reproducing and Stress-Testing Two Approaches to LLM Reasoning Reliability: Test-Time Probability Aggregation and Logic-Representation Editing
Minhan Cho and Jimin Kweon
We independently reproduce two recent methods for making large
language model (LLM)
reasoning more reliable, and stress-test them across domains and models (RPC across four new task domains with Qwen3-8B, LCF across four 7-8B models). The first, RPC, aggregates
token probabilities and self-consistency at inference; the second, LCF, trains projectors that split hidden states into "content" and "logic" and edits the logic part toward a valid region. Validating such reliability claims matters because the original evaluations are run by each method's own authors and were never independently reproduced or stress-tested across models and domains, and LCF shipped no public code. We re-run RPC's published-path aggregation and re-implement LCF's projector, contrastive, and intervention pipeline, then extend both to text-to-SQL, legal extraction, fallacy identification, and precedent grading, and probe LCF's representation directly. RPC reproduces the original grid exactly on the authors' released
reasoning paths; on four new domains its edge over self-consistency is never significant (ties or small mixed differences, paired p >= 0.28), and on BIRD, the one domain where we vary the budget, the edge grows with K as predicted but its largest gap (+2.5 accuracy at K=32, p=0.16) reverses to -0.25 when we enlarge the sample to n=200. LCF's logic-validity direction is real but weak (0.82 separability at the single best sub-layer versus 0.95 for a semantic-attribute control); its one positive effect (Qwen3 $\Delta$Prob) is not significant (p=0.56), while it significantly reduces $\Delta$Prob on two of the other three models.
Read the full abstract ▾
Artificial Intelligence
Time Present and Time Past: Benchmarking Large Language Models on Temporally Evolving Document Understanding
Mahbub E Sobhani, Md. Faiyaz Abdullah Sayeedi, Fahmid Hasan Chowdhury et al. (+4 more)
Evolving documents, such as laws, tax codes, and software documentation, are amended, replaced, and sometimes reverted over time, so a question has different correct answers at different dates. In contrast to encyclopedic knowledge, where an old fact is simply overwritten, an amendment is itself an official text that states what it replaces and when it takes effect, and the earlier version stays correct for its validity period. The central challenge is therefore version resolution, that is, identifying the version in force on the queried date. Existing temporal QA
datasets treat time only as an annotation, so version resolution stays untested. We present TIDE, an expert-verified
benchmark of 3,050 QA pairs over 644 official customs instruments issued between 1969 and 2025 by the Government of Bangladesh, covering eight task types over deeply code-mixed documents that are heterogeneous in layout and dated in two calendars. In addition, we evaluate nine recent LLMs under a single protocol across parametric, gold-context, and
retrieval access, scored by a three-judge LLM council with a hard date gate separating correct meaning from correct time. The best macro-averaged accuracy is only 68.5%. Resolving a version from an implicit date reaches 59.7%, and detecting that the supplied version does not govern the query reaches only 26.7%. Models are more likely to find correct versions than to reject incorrect ones, and they tend to follow a confident parametric answer over the supplied authoritative text. All code and data are available at https://github.com/icsetepa44/TIDE
Read the full abstract ▾
All authors Mahbub E Sobhani, Md. Faiyaz Abdullah Sayeedi, Fahmid Hasan Chowdhury, Md Adnan Arefeen, Farig Sadeque, Md. Faizul Bari, Swakkhar Shatabda
Artificial Intelligence
Understanding Calibration and Truncation Error Propagation in Training-Free Low-Rank Compression for LLMs
Mohanad Odema, Gabrielle De Micheli, Dayin Gou et al. (+3 more)
Training-free low-rank compression frameworks have been gaining prominence for LLM compression given their effectiveness in reducing model parameter count while maintaining task-level accuracy. However, existing SOTA frameworks share two key limitations: (1) residual errors in calibration data activations accumulate across layers during compression, causing misalignment between representations simulated at compression time and those experienced at inference; (2) the assumption that layer importance distribution is preserved post-compression does not hold. Together, these two effects introduce misalignment in the compression process in relation to the deployed model. We study these effects and propose a simple, training-free methodology compatible with existing frameworks to mitigate them, comprising: (1) Layer-by-Layer Compression with Calibration Correction; (2) Iterative Compression with Rank Allocation Correction. Implemented atop an existing SOTA decomposition framework, and evaluated on Llama and Qwen3 models across various
benchmarks and compression rates, our approach demonstrates up to ~1-2.5 accuracy point improvements over per-weight and joint decomposition baselines on zero-shot tasks.
Read the full abstract ▾
All authors Mohanad Odema, Gabrielle De Micheli, Dayin Gou, Nilesh Malpeddi, Prathamesh Vaste, Jacob Song
Artificial Intelligence
MathShikkha: A Controlled Study of Answer-Only and Chain-of-Thought Supervision for Bangla Mathematical Reasoning in Small Language Models
Rahma Simin Ali and Jawad Hossain
Mathematical
reasoning remains challenging in low-resource
languages such as Bangla. We study whether teacher-generated Bangla Chain-of-Thought (CoT) supervision provides benefits beyond ordinary supervised fine-tuning. We construct MathShikkha, a Bangla mathematical
reasoning dataset with GPT-5.4-generated rationales, and fine-tune four 4B–7B student models under a matched protocol in which answer-only and CoT conditions share data splits, response-only loss masking, decoding, and scoring, differing only in the training target. In-domain, CoT provides no significant improvement over answer-only fine-tuning for three stronger backbones (paired bootstrap 95% CIs include zero; exact McNemar $p \geq 0.17$), despite generating 15–52$\times$ more
tokens, but significantly improves the weaker 4B model by 18.56 points ($p < 0.0001$). On the larger, contamination-audited BanglaMATH
benchmark, this pattern reverses: CoT significantly outperforms answer-only supervision for all four models by 20.1–28.1 points (all $p < 0.0001$). Answer-only fine-tuning also reduces out-of-domain accuracy below the base model for three models, whereas CoT preserves or improves it for all four. A human study with two co-author annotators, external-expert adjudication, and Cohen's $\kappa = 0.76$–$1.00$ finds no significant CoT improvement over the base model on
reasoning-content criteria; instead, its measurable effect is target-
language adherence and producing inspectable
reasoning. Overall, rationale supervision's value depends on backbone capability and distribution shift: in this setting, its main benefits are Bangla adherence, auditable
reasoning, and out-of-domain robustness rather than improved in-domain
reasoning validity.
Read the full abstract ▾
Artificial Intelligence
TrustRoboReward: Preference-Ordered Isotonic Score Editing for Multi-Paradigm Robot Reward Models
Yidong Wang, Yan Zhan, Ziteng Feng et al. (+16 more)
Reward models are a bottleneck for reinforcement learning in
embodied AI. Long-horizon robotic manipulation requires scalable
vision feedback beyond handcrafted rewards or task-specific annotations. Existing open-source VLM reward judges like RoboReward adopt simple 1–5 trajectory progress scoring, lacking pairwise preferences for RLHF, DPO and Bradley-Terry frameworks, while failing to optimize
video scene understanding. Augmenting RoboReward with pairwise comparison and
video-QA supervision causes inconsistency between pairwise preferences and pointwise scores, introducing training noise and hurting downstream performance—an issue aggregation methods such as TrustJudge cannot resolve. To address this, we propose TrustRoboReward, a multi-paradigm reward modeling framework equipped with Preference-Ordered Isotonic Score Editing (POISE). We construct a unified four-paradigm
dataset with trajectory progress scoring (Score-A),
video-QA answer quality scoring (Score-B), and their pairwise counterparts (Pair-A, Pair-B). Pairwise labels align better with human judgment than pointwise scores, inspiring us to calibrate pointwise scores to avoid score-pair reversals against pairwise preferences. POISE rectifies pointwise scores and eliminates cross-paradigm reversal conflicts unresolved by TrustJudge. Theoretically, POISE reduces score-pair reversal conflicts from 20.15% to 0%, whereas TrustJudge retains 20.46% conflicts on the same corpus. Evaluated on our
benchmark, Qwen3-VL-4B trained with POISE achieves an overall reward score of 77.96%, nearly matching GPT-5-mini (78.09%, gap 0.13%) and outperforming the strongest RoboReward-4B baseline by 10.13%. It also lifts test-time score-pair consistency to 71.90%, exceeding RoboReward-4B (57.26%) and GPT-5-mini (68.09%). Integrating TrustJudge aggregation during inference boosts the overall score to 78.57%, surpassing the GPT-5-mini teacher model.
Read the full abstract ▾
All authors Yidong Wang, Yan Zhan, Ziteng Feng, Zhenyu Cui, Ziyi Zhou, Renzhao Liang, Jiaxuan Zhu, Zilei Yang, Yiran Zhao, Zhongkuan Mao, Bo Jia, Hanchu Ni, Chenggang Xie, Biao Liu, Yi Zhang, Yong Dai, Xiaozhu Ju, Wei Ye, Shikun Zhang
Artificial Intelligence
HoloAegis: Frozen Representation, Topological Inference — Minimally Parametric Safety Manifolds and Their Capability Boundaries for LLM Guardrails
Tak Ho Alex Li, Kaijie Liu, Lik-Hang Lee et al. (+3 more)
Current LLM safety guardrails face a fundamental tension: fine-tuning distorts pre-trained representations while generative judges incur prohibitive inference costs. We ask a complementary question: how far can safety be achieved through pure
geometric reasoning over frozen representations, and where does it fail? We present HoloAegis, a minimally parametric topological inference framework that decouples representation from
reasoning: an un-fine-tuned encoder maps text to the unit sphere S^{d-1}, and all decisions reduce to Gibbs-Boltzmann free-energy differences over pre-computed anchor centroids. We contribute a boundary-mapping study rather than a leaderboard claim. On a frozen three-
benchmark protocol, HoloAegis (3.2 MB) statistically matches WildGuard-7B (14 GB) on toxicity (0.96 vs. 0.96), exceeds it on harmful behaviors (0.99 vs. 0.79), and cedes oversafety detection (0.62 vs. 0.98) – while ShieldGemma-2B fails on indirect harms (0.34). These failure modes are complementary and mechanistically traceable: potential-difference scoring senses
manifold clustering, whereas policy-conditioned LLM judging requires explicit taxonomy matching. We restate our Topological Boundary Stability
conjecture in ratio form and validate it via reference-set bootstrap: anchor banks reduce score variance 4-15x and boundary displacement to approximately 0.44 + 0.23 sqrt(k/K) of the full-space
estimator. Per-domain analysis further reveals that
geometric separability tracks within-domain semantic homogeneity. Our results chart where
geometric guardrails substitute for, and where they must defer to, LLM judges.
Read the full abstract ▾
All authors Tak Ho Alex Li, Kaijie Liu, Lik-Hang Lee, Kin Chung Ho, Ping Shum, Michael K. Ng
Artificial Intelligence · Cross-listed from cs.NE
SuperNeuroMAT: An Efficient Matrix-based Simulator for Spiking Neural Networks
Prasanna Date, Kevin Zhu, Shruti Kulkarni et al. (+12 more)
Spiking neural networks (SNNs) offer a promising pathway to energy-efficient AI and
brain-inspired computing. However, their widespread adoption is hindered by a lack of fast, accessible, and versatile simulation frameworks. In this paper, we introduce SuperNeuroMAT, an open-source, scalable, and highly efficient Python-based SNN simulator. We devise a novel
matrix-based approach to model the leaky integrate-and-fire (LIF)
neuron dynamics and natively support dense and sparse execution modes. This enables fast simulation of approximately 10,000
neurons in dense mode and 100,000
neurons in sparse mode on standard laptops and desktops without requiring specialized hardware. We demonstrate that SuperNeuroMAT consistently outperforms four established SNN simulators—NEST, Brian2, BindsNET, and snnTorch—on two performance metrics (execution speed and peak resident
memory) and across various network sizes and connection probabilities. Furthermore, we demonstrate SuperNeuroMAT's applicability across a diverse set of problems. SuperNeuroMAT can efficiently handle conventional machine learning
benchmarks such as the Digits and citation network
datasets as well as neuromorphic event-based
vision tasks such as N-CARS and ASL-DVS. Moreover, it can be extended beyond machine learning workloads and facilitate general-purpose workloads. We validated this by implementing the neuromorphic shortest path
algorithm and two arithmetic primitives (addition and multiplication). SuperNeuroMAT can be installed via the Python Package Index (PyPI), thereby lowering the barrier to entry into the field of neuromorphic computing and accelerating the broader development of neuromorphic
algorithms.
Read the full abstract ▾
All authors Prasanna Date, Kevin Zhu, Shruti Kulkarni, Ashish Gautam, Chathika Gunaratne, Robert Patton, Tyler Nitzsche, Ian Mulet, Zachary Johnson-Scott, Addison Helms, Duncan Rowden, Simon Weston, Maryam Parsa, Catherine Schuman, Thomas Potok
Artificial Intelligence
Yesterday's Shield, Today's Spear: A Self-Evolving Safety Guardrail in Production
Cong Ming, Jingyi Chen, Bin Liu et al. (+4 more)
Deployed LLM safety guardrails are predominantly static: trained once and frozen at release, while new jailbreak techniques and previously un-addressed harmful categories emerge within days, leaving the defense perpetually a step behind. We present SESG (Self-Evolving Safety Guardrails), a multi-
agent system running in production. SESG monitors the live traffic behind a deployed guardrail and surfaces two classes of failure: jailbreaks novel in form and harmful categories novel in content. Once a failure is confirmed, a generation
agent synthesizes paired training data targeted at it; a validation
agent rebalances the batch toward the direction in which the deployed model errs, so that the model's own mistakes steer its training set; and a routing
agent matches the training action to the diagnosed gap and returns the next version to production. Over six rounds of live
evolution (V0 to V6), a 1.7B guardrail adapts to a new threat in 16-24 hours, with about 2 hours of human effort, versus the 40-90 hours of the manual process it replaces. On six emerging threats, it outperforms static guardrails from 0.6B to 9B and an adaptive baseline while preserving its general screening competence. Since April 2026, SESG has been the primary update pipeline of Sangfor's guardrail, autonomously closing 14 of 15 new threat scenarios in two months. We release 9 test sets for the 6 new threats at https://github.com/Trams1017/SESG. Warning: This paper contains examples that may be harmful or offensive.
Read the full abstract ▾
All authors Cong Ming, Jingyi Chen, Bin Liu, Qi Chu, Tao Gong, Nenghai Yu, Yingfei Xiang
Artificial Intelligence
Aero Realtime: Fully Aligned Input-Output Streams for Low-Latency Streaming Multimodal Generation
Kaichen Zhang, Wei Huang, Keming Wu et al. (+2 more)
Existing streaming
multimodal models process observations incrementally but still follow a turn-based prefill-then-decode pattern, making them non-duplex: new observations cannot naturally enter an active generation stream. Proactive alternatives use micro-turn polling or external response gates, which fragment continuous interaction, decouple response timing from
language generation, and complicate KV-cache-friendly serving. We introduce Aero Realtime, a 4B streaming
multimodal model with a duplex architecture for realtime generation. Aero Realtime aligns
video, audio, and textual output on a shared temporal grid, where each approximately 80-ms audio slot predicts either a lexical
token or a silence
token. This allows input and output to advance together, enabling one autoregressive objective to learn both when to respond and what to generate. During inference, Aero Realtime appends only the newest
multimodal slot, carries forward the previous output state, and reuses the KV cache for efficient incremental execution. We further provide a complete training and serving recipe, including realtime QA construction, slot-aligned supervision, hardware-aware distributed training, and resumable inference. On four NVIDIA A6000 workstation GPUs, Aero Realtime maintains 84-ms median and 173-ms P95 processing lag over 20 minutes of a continuously streamed
video, remaining within 200~ms of the source timeline. These results demonstrate the feasibility of fully aligned input-output modeling for duplex, proactive, and hardware-aligned
multimodal interaction.
Read the full abstract ▾
All authors Kaichen Zhang, Wei Huang, Keming Wu, Bo Li, Xiaojuan Qi
Artificial Intelligence · Cross-listed from cs.CR
SkillsMetric: Mapping the Detection Boundary of Static Analysis for Malicious Agent Skills
Xinze Chen, Chi Zhang, Ping Ji et al. (+1 more)
Agent Skills—structured packages of instructions and scripts that augment LLM-based
agents—are rapidly proliferating, yet their security properties remain under-explored. We present SkillsMetric, a five-stage static analysis framework that scores skill packages along pattern density, statistical anomaly, dataflow taint, import anomaly, and capability mismatch dimensions. We construct an
adversarial evaluation
dataset of 2{,}266 skills spanning 16~attack types across code-level, system-level, and semantic-level threats, and evaluate on the full SkillMD-138K corpus. Our framework achieves an AUC of 0.93 and 5-fold cross-validated F1 of 73.4%$\pm$0.5%, with strong detection of data exfiltration (93%) and steganographic payloads (93%). Crucially, we identify fundamental blind spots: host destruction attacks using common shell commands evade all five stages (0% detection), and
prompt injection via natural-
language manipulation achieves only 42% detection. These findings establish that static analysis alone is insufficient for skill security, motivating defense-in-depth architectures that combine fast static pre-screening with semantic review.
Read the full abstract ▾
All authors Xinze Chen, Chi Zhang, Ping Ji, Yimin Liu
Artificial Intelligence
LLM within MCP Matters: Measuring Inefficient Resource Utilization Driven by LLMs
Minhan Cho, Soyoung Park, Kihyeon Jeong et al. (+3 more)
The Model Context Protocol (MCP) standardizes how servers expose data and tools to Large
Language Models (LLMs). A common server design embeds frequently used reference data, such as identifier lookup tables, directly in the server instructions: the system-
prompt text a server hands to the host application. When a query concerns an entry of the embedded table, the model can act on it immediately instead of re-discovering the same information through a search tool. We test whether client LLMs actually consume such instruction-embedded data, reporting a 54,000-trial study across 24 LLMs (9 Claude, 6 Gemini, 9 GPT) on a production legal-information MCP server. A diagnostic condition that removes the competing search tool shows that failures are dominated by behavioral preference rather than missing capability. With search unavailable, 23 of 24 models read the embedded data reliably (hit ratio at least 98%); with a search tool merely present, 9 models drop below 15%. A 2^3 factorial analysis of three instruction-level interventions reveals strong interaction effects: combining all three restores at least 86% for 20 of 24 models, but individual interventions can backfire for specific model families. Per-server
prompt engineering is therefore a workaround rather than a fix; we argue that MCP host applications should provide an explicit mechanism that places server instructions ahead of tool selection in the client LLM's deliberation.
Read the full abstract ▾
All authors Minhan Cho, Soyoung Park, Kihyeon Jeong, Byeongkyu Jeon, Daejin Choi, Jinyoung Han
Artificial Intelligence
Hierarchical Self-Improvement: A Framework for Task-Specific Evolvable Agent Harnesses
Tailin Zhou
Modern LLM
agents are often improved by modifying
prompts, tools, or workflows manually, while the executable scaffold surrounding the model—the harness—is typically treated as a fixed artifact after deployment. This work studies an alternative where the harness is task-specific and continuously evolvable: each task family maintains its own harness, which is hot-swapped across iterations through a fixed task-injection seam and rewritten using environment feedback. We introduce Hierarchical Self-Improvement (HSI), a framework in which a single frozen LLM $M$ operates across three hierarchical scopes: a task harness $H$ that executes tasks, an evolver that rewrites $H$, and a meta-evolver that rewrites the evolver's strategy code under a frozen outer anchor. A thinking-on/off design isolates the contribution of harness
evolution by disabling
reasoning during task execution while enabling it during self-modification. HSI is bounded by two factors: a feedback-fidelity bound, since
evolution requires informative reward signals to guide selection, and a backbone capability bound, since harness redesign cannot overcome limitations of the frozen model. On BALROG with DeepSeek-V4-Flash-Preview as the frozen backbone, HSI achieves consistent gains over the initial harness on moderate-difficulty tasks ($+39.3$ on BabyAI, $+33.0$ on Crafter, $+25.0$ on TextWorld, and $+15.0$ on MiniHack, all in raw % Progress), while obtaining strong held-out generalization on BabaIsAI sub-suites ($0.98$ best-test on BreakStop and $1.00$ on GoTo from a $20\%$ unseen split). On tasks beyond the backbone's capability (NLE), harness
evolution provides no improvement. These results demonstrate task-specific harness
evolution as a viable axis for improving frozen LLM
agents under clear empirical limits. Code is available at https://github.com/TailinZhou/hsi.
Read the full abstract ▾