Artificial Intelligence · Cross-listed from cs.PL
Generative Compilation: On-the-Fly Compiler Feedback as AI Generates Code
Niels Mündler-Sasahara, Hristo Venev, Dawn Song et al. (+2 more)
Languages with rich static semantics, such as Rust, provide stronger guarantees for AI-generated code, but their strictness makes generation more difficult. Off-the-shelf compilers can provide useful feedback post-generation, but does not guide intermediate generation steps, such as those during autoregressive LLM decoding. Constrained decoding intervenes earlier by rejecting invalid
tokens during sampling, but requires white-box model access and costly reimplementation for semantic constraints. We introduce generative compilation, the first approach to obtaining compiler feedback on partial programs during generation. The core technical device is a sealor: a lightweight, mostly syntax-guided transformation that converts partial programs into complete ones that standard compilers can diagnose. It is designed such that possible-to-complete partial programs are never rejected, while preserving enough code context to catch genuine dead ends early. We construct such a sealor on a core Rust-like calculus and prove that it satisfies these properties, all mechanized in Lean. We extend it to the first partial-program checker for real Rust. We evaluate our method on challenging repository-level Rust coding tasks, across both frontier black-box and open-weight models. We show that generative compilation reduces non-compiling outputs and improves functional correctness, relative to standard post-generation feedback. It does so by detecting a broad range of errors close to their source and early during generation, thereby reducing errors cascades and enabling focused diagnostics. More broadly, generative compilation is a step toward making compilers a first-class citizen of AI-assisted programming active during generation, rather than a separate post-generation check.
Read the full abstract ▾
All authors Niels Mündler-Sasahara, Hristo Venev, Dawn Song, Martin Vechev, Jingxuan He
Artificial Intelligence · Cross-listed from cs.LG
Partially Correlated Verifier Cascades in LLM Harnesses: Concave Log-Odds, Polynomial Reliability, and Blind-Spot Ceilings
Jiangang Han
Serial verification gates are a core reliability primitive in LLM harnesses: a candidate answer is returned only if $k$ verifier calls all accept it. Under conditionally independent gates, the recent Odds Law (arXiv:2606.15712) shows that posterior log-odds grow linearly in $k$, so failure decays exponentially, and states that "a tight theory of partially correlated verifier cascades remains open." This note gives a minimal such theory. Modeling the per-instance false-accept rate on the generator's own errors as a latent variable $\alpha \sim G$ (de Finetti), the exact cascade posterior is $\ell_k = \ell_0 - \ln m_k$, with $m_k$ the $k$-th moment of $G$. Then: (i) $\ell_k$ is concave in $k$ for every non-degenerate $G$ – the Odds Law is its tangent at the first gate and an upper bound; (ii) for Beta$(a,b)$ latents, failure decays polynomially, $1-r_k \asymp k^{-b}$, with correlation parameter $\rho_v = 1/(a+b+1)$; (iii) a blind-spot atom of mass $1-\pi$ at $\alpha=1$ caps the evidence extractable from any number of gates at $-\ln(1-\pi)$ nats, so reliability saturates below 1; (iv) letting the true-accept rate also vary ($\beta \sim H$) yields a trichotomy – gates eventually always help, plateau, or actively harm – decided by the upper-tail exponents of $G$ and $H$, with closed-form crossover $k^\dagger$. The mechanism is survivorship: errors surviving gates are the high-$\alpha$ ones. The theory is measurable: $R$ repeated verdicts per instance identify the first $R$ moments of $G$, so two verdicts identify $\rho_v$; beta-binomial likelihood and NPMLE recover the reliability curve and the ill-posed ceiling. In synthetic tests, independence-based extrapolation underestimates failure by 20x at $k=5$ and ~3000x at $k=10$; the correlated fit at $R=8$ tracks held-out depths. The practical lever is decorrelation – changing model family, modality, or evidence source – not adding gates.
Read the full abstract ▾
Artificial Intelligence
AIMO Interpretability Challenge
Michal Štefánik, Philipp Mondorf, Andreas Waldis et al. (+15 more)
We propose the AIMO Interpretability Challenge, a competition on distinguishing robust from spurious
reasoning in frontier mathematical
language models based on the models' internal mechanisms. The challenge is motivated by a central limitation of standard
reasoning benchmarks: strong final-answer accuracy does not reveal whether a model relies on stable
reasoning mechanisms or exploits brittle
reasoning shortcuts. Building on AI Mathematical Olympiad (AIMO) problems and submissions, together with resources from the Fields Model Initiative, the competition will provide (1) newly-published olympiad-level math
reasoning problems and their symbolic representations, allowing generation of novel functional variants, (2) access to frontier
reasoning models, and (3) assessments of models'
adversarial robustness on these problems. Participants will use these resources, along with our computing infrastructure support, to develop methods for identifying which models solve problems robustly. Our competition will also create a new, open robustness
benchmark and baseline systems, aiming to provide a lasting foundation for standard benchmarking in mathematical
reasoning and interpretability. Scientifically, the competition connects interpretability and generalization research around a central question in AI research: can we determine if, and to what extent, the decision-making of frontier AI models is generalizable and thus, reliable?
Read the full abstract ▾
All authors Michal Štefánik, Philipp Mondorf, Andreas Waldis, Michal Spiegel, Robert Belanec, Chuan Yang, Maria Matveev, Jan Mazak, Qianying Liu, Markus Schepers, Josef Kuchař, Marek Kadlčík, Adam Vawda-Oomerjee, Chaoran Liu, Simon Frieder, Barbara Plank, Fazl Barez, Pontus Stenetorp
Artificial Intelligence
Experience Memory Graph: Extracting Reusable Correction Knowledge from Agent Trajectories
Wenjun Wang, Yuchen Fang, Fengrui Liu et al. (+2 more)
Large
Language Model (LLM)
agents have shown remarkable capabilities in autonomous decision-making by generating sequential trajectories of states, actions, and observations. However, in complex, long-horizon tasks, these
agents frequently suffer from compounding errors and struggle to recover from failures. Despite advances in experience
memory, extracting reliable and transferable correction knowledge from historical trajectories remains challenging. Existing methods, whether based on test-time reflection or offline
memory construction, largely rely on LLMs to diagnose failures and formulate corrective strategies, which can produce brittle or task-specific insights that are difficult to generalize. To address this challenge, we propose Experience
Memory Graph (EMG), a framework that reformulates
agent failure recovery as a graph matching problem. During offline construction on the training set, we convert both failed exploration trajectories and successful expert trajectories into directed action decision graphs. By matching these graphs, we extract common subgraphs and graph edit paths that explicitly indicate how to correct failures, and store them in a
memory graph with intra-task nodes and cross-task edges. At test time, EMG retrieves relevant insights and guides the
agent in a single test-time attempt without additional reflection-and-replay rounds. Experiments on ALFWorld and ScienceWorld demonstrate consistent improvements over the evaluated
memory construction and competitive baselines across seen and unseen splits.
Read the full abstract ▾
All authors Wenjun Wang, Yuchen Fang, Fengrui Liu, Zibo Liang, Kai Zheng
Artificial Intelligence · Cross-listed from stat.ME
Verifying formulas for interventional distributions
Francesco Freni, Leonard Henckel and Sebastian Weichwald
We formalize verification in causal graphical models: deciding whether a given observational formula identifies a target interventional distribution. This opens a problem complementary to identification, asking not whether any identifying formula exists, but whether the given formula is identifying. We show that even sound and complete solutions to identification do not solve verification. We propose a falsifier as a first practical route forward, prove that it induces an almost-surely correct verifier for regular exponential-family models, and use the resulting verifier to develop the gateway test, which finds all sets admissible for use in a front-door formula.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
Unleashing Multimodal Large Language Models for Training-free HOI Detection in the Wild
Ting Lei, Jialin Liu, Zhu Xu et al. (+2 more)
Human-object interaction detection (HOID) has traditionally been formulated as a supervised detection problem over predefined interaction categories. While such paradigms achieve strong performance on closed-set
benchmarks, they fundamentally entangle interaction understanding with
dataset-specific supervision, limiting their ability to generalize to open-world and compositional scenarios. Recent HOI detectors attempt to leverage MLLMs through prompting strategies to transfer interaction-specific knowledge. However, such
prompt-based approaches primarily focus on extracting discriminative representations from pretrained models, while underexploring their inherent
multimodal reasoning capabilities. As a result, they struggle to provide informative contextual
reasoning for ambiguous and open-world interaction scenarios. In this work, we present AgentHOI, a training-free, agentic framework that transfers the generalist
multimodal reasoning capabilities of foundation models to HOI detection in the wild. Instead of learning interaction classifiers, AgentHOI modularly orchestrates complementary
vision foundation modules to perform open-ended semantic
reasoning and spatial grounding in a coordinated manner. To address the challenges of incomplete interaction discovery and ambiguous localization in complex scenes, we introduce two key mechanisms: (1) Context-aware Multi-round
Reasoning, which progressively refines interaction hypotheses to ensure exhaustive and compositional HOI discovery, and (2) Multifaceted Interaction Localization, which enhances grounding precision by generating instance-specific descriptions that integrate semantic, spatial, and appearance cues. Extensive experiments demonstrate that AgentHOI achieves superior performance over state-of-the-art supervised and weakly supervised methods in real-world settings, despite requiring no HOID data for training.
Read the full abstract ▾
All authors Ting Lei, Jialin Liu, Zhu Xu, Yuxin Peng, Yang Liu
Artificial Intelligence · Cross-listed from cs.CY
AI-Augmented Human Resource Management? Insights from German companies
Yannick Kalff and Katharina Simbeck
This study examines the integration of AI into Human Resource Management in German companies. We ask if and how AI-based technologies are \enquote{augmenting} human resource management. Organisations employ generative AI or predictive analytics to transform traditional human resource functions, to streamline routine tasks and to reallocate resources toward strategic, people-centred activities. Our findings from interviews and group discussions and a survey (N=410) reveal that while AI tools enhance HR analytics capabilities, their adoption mainly serves efficiency and rationalising goals. The introduction of AI tools is shaped by organisational transformation factors such as digital infrastructure, co-determination frameworks, and ethical implications. The research highlights both the strategic potential for improved talent development and the challenges posed by data governance and algorithmic transparency. Overall, this work contributes to understanding the ambiguous role of technological change in HR, which promises to augment predictive capabilities yet serves the ends of efficiency and rationalisation.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
NodeImport: Imbalanced Node Classification with Node Importance Assessment
Nan Chen, Zemin Liu, Bryan Hooi et al. (+3 more)
In real-world applications, node classification on graphs often faces the challenge of class imbalance, where majority classes dominate training, resulting in biased model performance. Traditional GNNs often struggle in such scenarios, as they tend to overfit to majority classes while underrepresenting minority classes. Existing solutions, which either prioritize nodes based on class size or synthesize new nodes for minority classes, often fall short of effectively addressing this imbalance issue. This paper introduces an approach to class-imbalanced node classification by utilizing a balanced meta-set for importance measurement, where a training node is considered significant if it enhances model performance under an unbiased setting. Our method identifies important nodes that can counteract class imbalance and utilizes them for model training, allowing for fine-grained and dynamic node selection throughout the training process. We theoretically derive a formula to directly assess node importance, reducing computational overhead and providing an intuitive threshold for node selection. Guided by this metric, we develop a novel framework that filters valuable labeled, unlabeled, and synthetic nodes that enhance model performance in an unbiased context. A key advantage of this framework is its separation of the synthetic node generation process from the filtering process, ensuring compatibility with various node generation methods. Furthermore, we introduce a strategy to construct a high-quality meta-set that closely approximates the overall feature distribution, ensuring robust representation of each class. We evaluate our framework, NodeImport, across multiple
datasets using popular GNN architectures, demonstrating its superiority over existing baselines. Our results highlight the flexibility and effectiveness of the framework in mitigating class imbalance, leading to improved outcomes.
Read the full abstract ▾
All authors Nan Chen, Zemin Liu, Bryan Hooi, Bingsheng He, Jun Hu, Jia Chen
Artificial Intelligence · Cross-listed from cs.CV
Multimodal Assessment of Pancreatic Cancer Resectability Using Deep Learning
Vincent Ochs, Christoph Kuemmerli, Florentin Bieder et al. (+12 more)
Accurate determination of pancreatic ductal adenocarcinoma (PDAC) resectability relies on evaluating how the tumor interacts with major peripancreatic vessels on CT imaging, yet expert assessment often shows substantial variability. We introduce a fully automated
multimodal deep learning framework that jointly analyzes 3D contrast enhanced CT and structured clinical information to classify patients into the three National Comprehensive Cancer Network (NCCN) resectability categories (upfront resectable, borderline resectable, locally advanced). The approach uses a Swin-UNETR backbone to obtain anatomy aware image representations through auxiliary
segmentation of pancreas, tumor, and vascular structures. These features are fused with a compact clinical
embedding derived from 17 routinely collected variables and processed by a lightweight classification head. Model training is guided by a dynamic multitask objective that adapts the balance between
segmentation and classification based on current tumor Dice performance, promoting feature representations that remain both anatomically informed and discriminative.
Read the full abstract ▾
All authors Vincent Ochs, Christoph Kuemmerli, Florentin Bieder, Julia Wolleb, Joel L. Lavanchy, Julia Ruppel, Jan Liechti, Stephanie Taha-Mehlitz, Christian Andreas Nebiker, Beat Mueller, Giuseppe Kito Fusai, Joerg-Matthias Pollok, Anas Taha, Philippe C. Cattin, Sebastian Staubli
Artificial Intelligence · Cross-listed from cs.CR
Traffic-Aware Randomized Smoothing for LLM-Based Network Intrusion Detection
Zhenpeng Li
Large
language model (LLM)-based intrusion detection systems (IDS) are increasingly studied for security monitoring, yet their robustness against feasible traffic manipulation remains largely empirical. We present Traffic-Aware Randomized Smoothing (TA-RS), a classifier-agnostic certified defense that injects Gaussian noise exclusively into the directly controllable (DC) subspace – features a remote attacker can modify – during both fine-tuning and certification, aligning the smoothing distribution with the attacker-controllable subspace. We identify a critical prerequisite: applying standard randomized smoothing to clean-trained LLM-IDS yields weak certified accuracy in three of four (model,
dataset) pairs tested (14-33%, at or below random) and only 57% in the fourth (43 pp below the noise-augmented result); noise-augmented fine-tuning recovers to 68-100% on two of three
benchmark datasets (at sigma=0.25). At the L_inf-equivalent threshold R_inf = epsilon*sqrt(|DC|) (epsilon=0.05), TA-RS achieves 55-100% certified accuracy on CIC-IDS-2018 and HIKARI-2021, with median certified radii (R approx 0.45-0.96) exceeding R_inf by 1.8-5x (across sigma=0.25-1.00). Against a fairly trained iso-trained RS baseline the residual advantage is
dataset-dependent (4-19 pp on CIC-IDS-2018). The larger gap – up to 72 pp against an isotropic RS baseline that shares the DC-noise-augmented training recipe – primarily reflects the training-certification mismatch rather than DC
alignment alone: isotropic test-time noise perturbs uncontrollable features the attacker cannot exploit, triggering abstention rates up to 68%. RT-IoT2022 probes the limits of the method: it fails under the default fine-tuning recipe but recovers to 76%/69% certified accuracy (LLaMA3-8B/Qwen3-8B) when noise augmentation is increased.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.IT
Romie Banerjee
This paper establishes a direct analogue of the classical Coding
Theorem in the setting of
symmetry groups. We consider computable bijections on the set of binary strings and define the
symmetry prior of a string x as the probability that a randomly chosen
symmetry from a given group G has x as its unique fixed point. We show that for any fix-retractable
symmetry group G, a group admitting a computable section that selects an isolating
symmetry for every string, the
symmetry prior is a universal lower semi-computable semi-measure. In this case, the
Geometric Coding
Theorem holds. This result is a coding-theoretic restatement of a
theorem of Trejo, Kreinovich and Longpré, who showed that the complexity of describing a string by a
symmetry with that string as its unique fixed point equals its Kolmogorov complexity. Our contribution is to recast it in terms of algorithmic probability and to treat the
symmetry group as a parameter, identifying fix-retractability as exactly the condition under which
symmetry complexity collapses onto Kolmogorov complexity.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.AR
Kaleido: Algorithm-Hardware Co-Design for Video Diffusion Transformers by Exploiting Latent Space Correlations
Wenxuan Miao, Haosong Liu, Weiming Hu et al. (+9 more)
Video diffusion transformers (vDiTs) generate high quality
video but introduce extremely high compute cost due to the long
diffusion timesteps and self
attention computation. As
diffusion timesteps are reduced, the computation cost of self
attention becomes the dominant bottleneck. Existing acceleration approaches largely inherit sparse
attention techniques from large
language models, which fail to consider the unique spatiotemporal correlation of
video data. This paper presents Kaleido, an
algorithm hardware codesign that accelerates all operations in vDiTs by exploiting channel-wise spatiotemporal correlations in latent space. Based on this insight, we propose a lightweight channelwise reuse
algorithm that skips redundant computations by reusing partial results while preserving higher generative quality than prior methods (>17 dB). To efficiently support this
algorithm, we design a systolic array like accelerator with reconfigurable processing elements and a lightweight data dispatcher to mitigate irregular sparsity and data access patterns introduced by our reuse
algorithm. Evaluations across three mainstream vDiT models show that Kaleido achieves up to 5.9x speedup and 16.0x energy savings over state of the art accelerators.
Read the full abstract ▾
All authors Wenxuan Miao, Haosong Liu, Weiming Hu, Zihan Liu, Aiyue Chen, Jianlin Yu, Yiwu Yao, Yiming Gan, Jieru Zhao, Jingwen Leng, Minyi Guo, Yu Feng
Artificial Intelligence · Cross-listed from cs.LG
MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model
Charilaos Papaioannou, Ioannis Tsantilas, Dimitris Giannakakos et al. (+9 more)
Single-task fine-tuning of graph neural networks (GNNs) for power grid problems exhibits a systematic failure mode: models that achieve the lowest in-distribution error degrade the most under
topology shift. We term this
topology overfitting: the tendency of task-specific gradient signals to encode relational structure particular to the training
topologies rather than the underlying physics, causing models to fail on unseen grids despite strong in-distribution performance. To expose and address this failure mode, we introduce MxGPS (Multiplex GPS), a multiplex graph
transformer that runs K task-specialised GPS branches over a shared node encoder, jointly trained on Static State Estimation (SSE) and AC Power Flow (PF) via a self-supervised pre-training and multi-task fine-tuning protocol, with a cross-branch
attention module evaluated in ablation. The joint SSE+PF objective forces the shared encoder to simultaneously satisfy complementary gradient signals, preventing it from overfitting to
topology-specific relational structure. Under a 3-fold sliding-window cross-validation spanning four unseen
topologies (14-, 24-, 162-, and 300-bus), MxGPS attains 0% boundary violation rate (BVR) on all four zero-shot Power Flow
topologies. Critically, models with substantially lower in-distribution PF error degrade by 190% to 1400% under
topology shift, whereas MxGPS degrades by only 39%, an inversion that directly implicates
topology overfitting as the failure mechanism rather than insufficient model capacity. With only 1.6M parameters (12x fewer than the GridFM reference baseline), MxGPS demonstrates that multi-task joint training is a principled and parameter-efficient mechanism for
topology-agnostic generalisation in power grid foundation models.
Read the full abstract ▾
All authors Charilaos Papaioannou, Ioannis Tsantilas, Dimitris Giannakakos, Vasilis Michalakopoulos, Sotiris Pelekis, Vangelis Marinakis, Arsam Aryandoust, Antonello Monti, Ricardo J. Bessa, Perdo P. Vergara, Jochen Cremer, Elissaios Sarmas
Artificial Intelligence · Cross-listed from cs.CV
Ablation-Corrected Evaluation of Attribution Maps in Echocardiographic Ejection-Fraction Models
Hyunkyung Han and Min Jung Kim
Attribution maps for echocardiographic ejection-fraction models are evaluated by their overlap with an expert left-ventricular annotation, compared against a chance level that is computed from an area ratio rather than measured. We measure it. Two architectures trained on the same task attain overlap at 3.55 and 4.20 times measured chance, an eighteen percent difference a reader would take as the size of the gap between them. It is not. Replacing the annotated ventricle with a composition-matched surrogate changes one model's prediction six times more than replacing an equal-area control region and the other's twice, a factor of three; on sixty-two percent of cases for the second, neither the ventricle nor the attributed region moves the prediction at all. The reliability of this measurement, from two independent intervention batches, is 0.90 to 0.98. Overlap does not track the difference: scored against an ablation-derived label of which cases the model depends on the ventricle for, it reaches an area under the curve of 0.52 and 0.67. The design of the ablation also determines its answer, since ablating only the two annotated frames leaves the ventricle indistinguishable from a control region while ablating across all frames does not. A pediatric cohort reproduces all of these. We give the measured chance level, the symmetric-ablation protocol, the reliability estimate, and the ablation-prediction score as things to report alongside overlap.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
How Agents Ask for Permission: User Permissions for AI Agents, from Interfaces to Enforcement
Alexandra E. Michael and Franziska Roesner
As AI
agents grow in prevalence, so too do their risks to users, creating a need for robust and usable systems for
agent permissions. Many recent proposals implement product-level permissions, where the same security policy is applied for all users of a given
agent framework. Yet different users have varying needs, necessitating support for user-level permission policies in agentic AI systems. We survey 21 proposed user-level
agent permissions systems and construct a taxonomy of approaches to specifying, deriving, and enforcing user-level permissions. We further analyze and compare against five commercial
agents. We find that despite a wide diversity of proposals, sizable gaps remain in usability, transparency, and user control, particularly between academic and commercial
agents.
Read the full abstract ▾
Artificial Intelligence
CAVA: Canonical Action Verification and Attestation for Runtime Governance of Agentic AI Systems
Zexun Wang
Agentic AI systems increasingly act through heterogeneous runtimes: local coding hooks, SDK tools, browser automation, managed-
agent traces, API gateways, and workflow engines. A single operational act such as publishing code, changing identity state, moving money, or exporting data may therefore be represented by many incompatible runtime records. This makes a basic governance question difficult to answer: what action was actually approved, what evidence binds the approval to execution, and can an independent verifier reproduce the same action identity later? This paper presents Canonical Action Verification and Attestation (CAVA), a runtime-semantics layer for converting heterogeneous
agent activity into canonical runtime action objects. CAVA is positioned below Proof-Carrying
Agent Actions (PCAA): PCAA defines the deployer-owned route-review-prove governance process, while CAVA defines the stable action object that process governs. The paper formalizes canonical action identity, semantic pattern detection, approval binding, receipt integrity, runtime-portable projection, and optional attestation substrates. We study a reference implementation through a 96-seed, 384-variant
benchmark covering semantic equivalence, semantic separation, wrapper bypass, false-positive control, approval binding, receipt reproducibility, attestation tamper detection, runtime portability, semantic pattern detection, policy degradation, and Azure deployment drills. The contribution is a systems formulation of action-level canonicalization and policy-addressable semantic patterns as a necessary substrate for deployer-side AI governance.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
Groc-PO: Grounded Context Preference Optimization for Truthful Multimodal LLMs
Zhixiao Zheng, Zheren Fu, Zhiyuan Yao et al. (+3 more)
Despite the rapid progress of
Multimodal Large
Language Models (MLLMs), they still suffer from untruthfulness issues, such as visual
hallucinations, content fabrication, and unfaithful
reasoning, which substantially undermine their faithfulness and practical utility.
Alignment methods based on human preference, such as Direct Preference Optimization (DPO), have been widely adopted to address these issues. However,
multimodal reasoning errors often propagate across stages, and final-answer errors can often be traced to mistakes in early grounding stages, yet standard DPO typically applies preference optimization at the final-answer level. This credit-assignment challenge means that supervision for early grounding stages is indirect rather than stage-specific, making it difficult to suppress error propagation arising from grounding drift and context inconsistency. To address this, we propose Grounded Context Preference Optimization (Groc-PO), a grounded preference optimization framework for MLLMs. We further construct the Grounded Context Preference
Dataset (GCPD), organizing multi-stage preference samples around three stages of Object Grounding, Contextual Grounding, and Grounded
Reasoning, to capture the formation, integration, and utilization of grounded context. By introducing more explicit preference supervision over multiple grounded stages, Groc-PO strengthens context-dependent
reasoning and mitigates cross-stage error propagation. Extensive experiments show that, compared with standard DPO and other strong baselines, Groc-PO achieves improved performance in
hallucination mitigation, faithful
reasoning, and overall reliability, supporting the value of more explicit grounded supervision for trustworthy
multimodal reasoning.
Read the full abstract ▾
All authors Zhixiao Zheng, Zheren Fu, Zhiyuan Yao, Chunxiao Liu, Dongming Zhang, Zhendong Mao
Artificial Intelligence
AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
Kai Chen, Zichen Ding, Jiaye Ge et al. (+20 more)
As Large
Language Models (LLMs) evolve into autonomous
agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipelines remain highly fragmented and tightly coupled, hindering reproducibility and causing redundant engineering. To address this, we introduce AgentCompass, an open-source, lightweight, and extensible infrastructure for evaluating LLM-based
agents. AgentCompass organizes the evaluation process around three independent components, namely
Benchmark, Harness, and Environment, thereby enabling flexible configurations without requiring the reimplementation of complex execution logic. Furthermore, it features a fault-tolerant asynchronous runtime and comprehensive trajectory analysis tools to transparently diagnose nuanced failure modes like reward-hacking. Natively supporting over 20
benchmarks across five capability dimensions, AgentCompass provides the community with a scalable and reproducible infrastructure for advancing
agent research.
Read the full abstract ▾
All authors Kai Chen, Zichen Ding, Jiaye Ge, Shufan Jiang, Mo Li, Qingqiu Li, Zehao Li, Zonglin Li, Tianhao Liang, Shudong Liu, Zerun Ma, Zixin Shang, Wenhui Tian, Zun Wang, Liwei Wu, Zhenyu Wu, Jun Xu, Bowen Yang, Dingbo Yuan, Qi Zhang, Songyang Zhang, Peiheng Zhou, Dongsheng Zhu
Artificial Intelligence · Cross-listed from cs.MA
Social Simulations: from Agent-Based Modeling to Digital Twins
Erica Cau, Andrea Failla, Valentina Pansanella et al. (+1 more)
This book chapter covers the
evolution of social simulation from classical
agent-based models, in which
agents interact according to explicitly defined behavioral rules, to AI-enhanced simulations based on Large
Language Models and, ultimately, Social Digital Twins: high-fidelity, data-driven representations of real-world socio-technical systems. Along this trajectory, we discuss the main methodological foundations, applications, advantages, and limitations of each paradigm, highlighting the progressive shift from abstract models designed to investigate general social mechanisms toward increasingly realistic computational representations of specific social systems.
Read the full abstract ▾
All authors Erica Cau, Andrea Failla, Valentina Pansanella, Giulio Rossetti
Artificial Intelligence · Cross-listed from cs.CV
Barnamala: Parameter-Efficient Handwritten Devanagari Recognition at Benchmark Saturation
Ashish Thapa and Samrat Karki
We built a compact convolutional network (1.11 M parameters) for 46-class DHCD Devanagari recognition and reached 99.73%, the highest reported at 15.6x smaller than prior state-of-the-art. We have effectively reached the saturation point: every model tested, large teacher ensembles included, hits the same 11-error intrinsic floor. No configuration achieves a statistically clear win under exact McNemar tests with Wilson confidence intervals. Even without knowledge
distillation, our student matches the nearest large-model baseline (17.32 M parameters; McNemar $p = 0.345$). Outside of DHCD, zero-shot on CMATERdb digits gives 76.6% and fine-tuning reaches 97.8%; corruption robustness is also far better than large baselines (mean corruption accuracy 75.7% vs. 38.7%). All artifacts are at https://github.com/Ampixa/barnamala.
Read the full abstract ▾
Artificial Intelligence
When Bots Join the Team: Bot Adoption and the Institutional Fabric of Open-Source Software Projects
Yongren Shi and Wenyi Gong
AI
agents are joining human teams, raising a basic question: when an automated
agent becomes a regular participant, does group organization strengthen or weaken? We study this question in open-source software, where bots open pull requests, review code, and merge changes alongside people, leaving a public record of every interaction. Treating bots as participants rather than tools, we examine 2,991 GitHub projects for two years before and after each adopted its first bot. We measure three capabilities that institutional theory links to durable coordination - repeated engagement, social
memory, and role differentiation - and two outcomes: conflict cascades and output distinctiveness. Bot adoption is followed by more repeated collaboration, greater recognition of specific bots in discussion, fewer conflict cascades, and more distinctive outputs. These changes cluster around adoption rather than accumulating gradually. Because we lack an untreated comparison group, we interpret the results as precisely timed associations, not causal effects. Two patterns are difficult for alternative explanations to account for: capabilities predict outcomes according to their function - coordination versus differentiation - rather than whether humans or bots provide them, and human-side capabilities account for the bot-conflict association but not the bot-distinctiveness association. The findings are consistent with a specific interpretation: predictable, rule-based
agents can become part of a community's social infrastructure. The bot is the occasion; social organization is the mechanism.
Read the full abstract ▾
Artificial Intelligence
Explaining Reinforcement Learning Agents via Inductive Logic Programming
Celeste Veronese, Edoardo Zorzi, Daniele Meli et al. (+1 more)
Explainable Reinforcement Learning (XRL) seeks to make Reinforcement Learning (RL) policies more transparent and interpretable, a key requirement in safety-critical and human-centric scenarios. However, it is mostly based on user studies, thus targeting the needs of a specific audience and lacking shared evaluation metrics. On the other hand, logic-based approaches within eXplainable Artificial Intelligence (XAI) provide compact, human-readable abstractions of decision-making. However, the systematic quantification of the explainability degree of logical representations remains an open problem. This work aims to advance the state of the art in XRL by introducing objective and planning-oriented metrics for policy explainability in RL settings. At the same time, it contributes to the field of logic for XAI by providing a principled way to quantify the explainability of logical rules, moving beyond common-sense assessments and simple propositional fragments. We employ Inductive Logic Programming (ILP) to extract symbolic representations of RL policies and define a novel set of explainability metrics, including activation rate, feature coverage, syntactic distance and semantic distance. These metrics quantify
alignment between symbolic rules and
agent behavior, the role of features in decision-making, and the
evolution of policies during training and across
agents in single and multi-
agent RL. Experiments across different RL domains show that the proposed metrics highlight action-specific learning dynamics beyond global return, provide fine-grained insights into domain features beyond classical approaches for global feature importance estimation, and uncover coordination, specialization, and adaptation patterns in MARL. Moreover, they provide crucial insights for the transfer and generalization of action-specific policies.
Read the full abstract ▾
All authors Celeste Veronese, Edoardo Zorzi, Daniele Meli, Alessandro Farinelli
Artificial Intelligence · Cross-listed from cs.CV
Beyond Color Geometry: Evaluating Human-Like Color Representations in Vision Models
Ayan Igali and Pakizar Shamoi
Do
vision models see colors the way humans do? Existing evaluations of color representations usually compare them with
geometric spaces such as CIELAB or with discrete color labels. These references capture perceptual distance or category membership, but not the graded way in which people organize colors. We evaluate color grounding against a fuzzy perceptual model with 86 graded categories fitted to human survey data. The framework can be applied to any image encoder and measures three complementary properties: category boundaries, category compactness, and graded
alignment beyond what color geometry alone can explain. Across eleven
Vision Transformer encoders, the category-level results are broadly similar, whereas graded
alignment differs substantially. Masked Autoencoders achieve the strongest beyond-geometry
alignment, with confidence intervals that do not overlap those of the other encoders. A layer-wise analysis further shows that masked reconstruction preserves this structure toward the output. On natural images, MAE represents surface color globally, while
language-supervised models encode color more strongly in relation to the foreground object. These results show that human-like color grounding has several distinct aspects that should not be reduced to a single score.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
Human4K: A Large-Scale 4K Multi-View Mocap Dataset for Whole-Body 3D Human Reconstruction
Tianshun Han, Ziyu Shi, Lijian Liu et al. (+7 more)
Recent advances in 3D human reconstruction have improved overall performance, yet current models still fail in the most challenging real-world scenarios. They often produce unstable geometry, inaccurate limb articulation and unreliable predictions under depth ambiguity or self-occlusion. A key reason is that existing
datasets still lack the combination of high-resolution images, high-precision annotations and diverse whole-body motions required to support robust reconstruction. To address this gap, we present Human4K, a large-scale 4K multi-view whole-body human reconstruction
dataset with mocap-accurate SMPL-X annotations. Human4K contains over six million 4K images captured by an eight-view high-resolution camera system synchronized with a professional Vicon motion capture setup, covering 11 subjects performing complex, highly articulated and strongly self-occluded full-body motions. All sequences are processed by a Motion-Retargeting and Refinement Module (MRRM) to ensure precise
alignment for the full body and extremities. Experimental results show that training with Human4K consistently improves whole-body reconstruction on standard
benchmarks, with particularly large gains for hands, feet and depth-ambiguous limb configurations.
Read the full abstract ▾
All authors Tianshun Han, Ziyu Shi, Lijian Liu, Ajian Liu, Benjia Zhou, Hugo Jair Escalante, Yanyan Liang, Sergio Escalera, Zhen Lei, Jun Wan
Artificial Intelligence · Cross-listed from cs.LG
Consensus as Privileged Context for Label-Free Self-Distillation
John Gkountouras, Josip Jukić and Ivan Titov
Sampling multiple solutions and returning the majority answer is among the most reliable ways to improve the
reasoning accuracy of large
language models without labels, and a growing family of methods converts this consensus signal into training supervision. However, existing approaches use consensus only in restricted forms: as a filter that selects solutions for fine-tuning, as a preference between answers, or as a scalar reward for reinforcement learning, discarding most of the information that the agreeing solutions contain. We present CANON (Consensus-ANchored self-
distillatiON), a label-free training method that turns consensus into dense,
token-level supervision. For each unlabeled
prompt, CANON samples multiple solutions, extracts the majority answer, and conditions a frozen snapshot of the model on a solution that reaches it; this consensus-anchored teacher then supervises the model on its own rollouts at every
token. Experiments on mathematical and scientific
reasoning benchmarks show that CANON improves pass@1 by up to 12 points, outperforming label-free reinforcement learning by 6 points at a seventh of its compute and approaching a teacher conditioned on gold solutions; trained on pooled unlabeled data, it transfers to held-out
benchmarks, matching training methods that use gold labels. Analysis suggests that the improvements are not pure distribution sharpening: after training, the model solves problems it previously never solved in 32 attempts, and its majority vote itself becomes more accurate.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
OvisOCR2 Technical Report
Shiyin Lu, Yinglun Li, Yu Xia et al. (+10 more)
We introduce OvisOCR2, a 0.8B document parsing model. OvisOCR2 is designed as an end-to-end parser: given a document page image, it generates a Markdown representation in natural reading order, covering text, formulas, tables, and visual regions. We build a data engine that combines filtered real-document annotations with synthetic pages whose rendered images and Markdown targets are derived from the same HTML source. The training recipe includes supervised fine-tuning, reinforcement learning on a 4B branch with a multi-component reward design, on-policy
distillation into the 0.8B model, and model fusion. On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing. On PureDocBench, OvisOCR2 also achieves the highest Avg3 score of 75.06. Beyond these two public
benchmarks, we evaluate OvisOCR2 on an in-house
benchmark designed to cover a broader set of long-tail and challenging scenarios. OvisOCR2 obtains the best overall performance among the compared methods, providing further evidence of its generalization and robustness. OvisOCR2 is available at https://huggingface.co/ATH-MaaS/OvisOCR2.
Read the full abstract ▾
All authors Shiyin Lu, Yinglun Li, Yu Xia, Yuhui Chen, An-Yang Ji, Jun-Peng Jiang, Qing-Guo Chen, Jianshan Zhao, En Lin, Haijun Li, Cheng Qin, Zhao Xu, Weihua Luo
Artificial Intelligence · Cross-listed from cs.RO
Jose Martínez-Fajardo, Pablo Pueyo, Fernando Caballero et al. (+1 more)
Natural
language interaction provides an intuitive way for non-expert users to communicate with robotic platforms. However, transforming user requests into executable
navigation actions remains a challenging task, requiring the integration of
language understanding, environment perception, and autonomous
navigation. This work presents a
language-driven
navigation framework that enables mobile
robots to interpret user requests in natural
language to move the
robot to a destination and autonomously navigate towards it. The framework is composed of modular ROS 2 components that cooperate to transform natural
language instructions into
navigation actions. Given a natural
language request referring to a target in the environment (e.g., "go to the mail box"), the system identifies the referenced object, estimates its position using RGB-D data, and generates a
navigation goal, which is then executed through the ROS 2 Nav2
navigation stack. The ROS 2-based implementation facilitates portability across different robotic platforms, requiring only the configuration of the corresponding topics and services. The system is evaluated in both simulation and real-world scenarios using a TurtleBot3 Waffle and a Unitree Go2
robot with a RealSense camera. Experimental results show that the framework successfully interprets both direct commands and contextual requests, generates meaningful natural-
language feedback, and navigates towards the desired target. These results demonstrate the feasibility of combining semantic perception and autonomous
navigation to provide an intuitive human-
robot interaction paradigm. Code will be released as open source upon acceptance.
Read the full abstract ▾
All authors Jose Martínez-Fajardo, Pablo Pueyo, Fernando Caballero, Luis Merino
Artificial Intelligence
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
Kun Yu, Jianhua Yang, Yixiang Chen et al. (+7 more)
Language-guided human following is an important capability for
embodied agents, but existing
benchmarks typically assume that the target person is visible at the start of an episode. This setting simplifies the problem and overlooks a more realistic requirement: an
agent often needs to first find a
language-described target and then persistently follow that target in a dynamic environment. While recent work has started to study human search, existing settings are typically evaluated in task-specific scenarios and often rely on stronger prior knowledge of the environment. Moreover, they usually treat searching and following as separate tasks and still lack a unified
benchmark for systematic evaluation. To address these limitations, we introduce the Unified
Embodied Seeking and Following
Benchmark (UESF-Bench), a large-scale and diverse
benchmark for
embodied human seeking and following. The
benchmark requires
agents to handle semantic-guided exploration, reliable behavior switching and recovery, and delayed identity grounding. To this end, we propose SeekFollow-VLA, a
vision-
language-action framework with a task-driven routing mechanism for latent phase inference and transition modeling between seeking and following. Experimental results show that SeekFollow-VLA achieves clear improvements over both single-head and dual-head baselines across single-person and multi-person environments, establishing a baseline for unified
embodied seek-and-follow.
Read the full abstract ▾
All authors Kun Yu, Jianhua Yang, Yixiang Chen, Changwei Wang, Hongyuan Yu, Yan Huang, Fushuo Huo, Ya Jing, Zhumin Chen, Keji He
Artificial Intelligence
STOCKTAKE: Measuring the Gap Between Perception and Action in LLM Agents with a Fair Oracle
Sagar Deb and Ashwanth Krishnan
LLM
agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the final cost cannot say why an
agent failed: it may have misread the world, or read it correctly and still failed to act (the knowing-doing gap). Existing evaluations cannot separate these two failures; their reference policies either read privileged information the
agent never sees, or are missing altogether. We introduce STOCKTAKE, a 26-week supply-chain replenishment
benchmark built as a factored partially observable Markov decision process with six hidden factor processes, designed so that a fair reference policy is computable: an exact Bayes filter per factor drives a rollout policy on the identical observation stream the
agent receives. Scoring each run between a symptom-blind base-stock floor (0) and this oracle (1) yields a skill score, and grading each week's written rationale yields a stated-belief detection lag and a knowing-doing rate, so state estimation and control are measured separately. On fifty seeds with curated stress profiles, Claude Sonnet 5, GPT-5.4, DeepSeek-V4-Pro, and Grok 4.5 detect 84-88% of hidden failures, typically within a week of onset, yet span skill scores from 0.62 to -0.23: two of the four end below the symptom-blind floor while naming factors slightly faster than the two that beat it. The failure has two faces. Where stress persists, 34-43% of correctly diagnosed stress weeks still end in stockout for every model, a rate that partly reflects the severity of the weeks models notice. That rate also runs opposite to skill: the two models under the floor stock out least on diagnosed weeks, so under-response is only one face of the gap, and their traces point to the other, responses whose cost exceeds what they protect. STOCKTAKE measures both directions of that failure.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
The SIGReg Objective as Variational Free Energy: A Theoretical Active-Inference Account of JEPA World Models
Fabio Arnez and Alexandra Gomez-Villa
Joint-
Embedding Predictive Architectures (JEPAs) are the dominant design for latent world models, yet they are usually justified by empirical performance rather than a normative principle. We show that the choice of anti-collapse regulariser determines whether a JEPA's training objective, a prediction loss plus a weighted
embedding regulariser, is a valid Active Inference (AIF) variational free energy. We organise four non-contrastive regularisers (VICReg, LogDet, PairDist, and SIGReg) into an
entropy-
estimator hierarchy indexed by a prior-miscalibration gap, and show that the gap's sign, whether the
estimator bounds the latent
entropy from above or below, decides whether the AIF surprise bound survives: VICReg and LogDet are unsafe upper bounds, PairDist a safe lower bound, and SIGReg eliminates the gap. We then prove a correspondence
theorem: under the standard constant-noise encoder model and successful SIGReg enforcement (isotropic-Gaussian
embeddings), the gap vanishes, the objective becomes an exact information bottleneck, the surprise bound is preserved, and the latent goal cost becomes an exact proxy for AIF pragmatic value, whereas VICReg leaves an irreducible second-order anisotropy term. We extend the correspondence to multi-step expected free energy, ensemble epistemic value, and a learned-policy regime, and we identify the one AIF term no current JEPA world model computes: the state-epistemic value, a future-state coverage signal. The predictions differ in kind, not degree, and are stated here as theoretical consequences left for empirical test in separate work; full proofs are in Appendix A, and the algebraic core of every result is machine-verified in Lean 4 (Appendix D).
Read the full abstract ▾