Artificial Intelligence
Lose the Order, Keep the Hierarchy: Deordering HTN Plans
Takudzwa Togarepi, Gaspard Quenard, Damien Pellier et al. (+1 more)
Hierarchical Task Network (HTN) planning is a powerful planning formalism based on task decomposition. Although most of the literature studied plan generation, comparatively less
attention has been paid to post-plan optimization. In particular, plan deordering has been extensively studied in classical planning but remains under-researched in the HTN setting. Plan deordering removes unnecessary ordering constraints between actions in a plan whilst keeping the plan valid. In this paper, we adapt two established plan deordering techniques from classical planning by extending the techniques to account for hierarchical decomposition constraints. We evaluate our proposed approaches on the IPC 2023 Partial-Order HTN
benchmarks and we compare them against Optiplan, an HTN planner that generates partially ordered plans directly. Our results show a substantial reduction in number of ordering constraints in both our implementations. Although we also observe a reduction in critical path length, the improvements are less pronounced.
Read the full abstract ▾
All authors Takudzwa Togarepi, Gaspard Quenard, Damien Pellier, Humbert Fiorino
Databases
Property Testing for Recursive Query Languages
Isolde Adler, Carsten Lutz, Quentin Manière et al. (+2 more)
In the context of database querying, property testing provides a framework for testing query answers with high confidence while inspecting only a sublinear part of the database, through completion queries and size queries. A fundamental result of Chen and Yoshida (2019) states that non-satisfaction of a Boolean conjunctive query $q$ is testable with a constant number of such queries and one-sided error if and only if $q$ is equivalent to an $\alpha$-acyclic query. In this article, we initiate the study of property testing for recursive query
languages, focusing on two-way regular path queries (2RPQs) and monadic Datalog. One of our main results is positive: non-answers to any 2RPQ are constant query testable with one-sided error. We extend this slightly to a certain class of monadic Datalog programs in which recursion is restricted to be linear and rule bodies must be $\alpha$-acyclic. Turning towards unrestricted monadic Datalog, we next show that if a monadic Datalog program $\Pi$ is not equivalent to an $\alpha$-acyclic program, then falsity of $\Pi$ is not constant query testable with one-sided error. This is under the assumption that all rule-bodies are self-join free. We leave open the case of monadic Datalog programs with $\alpha$-acyclic rule bodies that are not restricted to linear recursion, but observe as a first step that there exist $\alpha$-acyclic programs that are mildly non-linear and constant query testable with one-sided error.
Read the full abstract ▾
All authors Isolde Adler, Carsten Lutz, Quentin Manière, Marcin Przybyłko, Lukas Schulze
Robotics
Revisiting Topological Graphs for Macro Action based Closed-loop Reinforcement Learning of Vision Language Navigation in Continuous Environment
Shuhao Ye, Sitong Mao, Yuxiang Cui et al. (+7 more)
Vision-
Language Navigation in Continuous Environments (VLN-CE) requires an
agent to follow natural
language instructions through unseen environments. Existing imitation learning (IL) pipelines struggle in this closed-loop setting: behavior cloning suffers from distribution shift, and DAgger's expert actions become ambiguous upon trajectory deviation. While Reinforcement Learning (RL) offers a natural paradigm to address this, directly applying RL to micro action spaces is sample-inefficient due to reward sparsity. To overcome this bottleneck, we reformulate VLN-CE as a Hierarchical Markov Decision Process (MDP), explicitly decoupling high-level planning from low-level control. By abstracting the environment into a topological graph, our high-level policy operates on a macro action space of frontier nodes, with a training-free low-level controller acting as its state transition, which significantly compresses the decision horizon and makes closed-loop RL tractable. To support RL optimization on the macro MDP, we propose an action-aware value head to effectively evaluate state values under the dynamic frontier action space, powering a graph-based PPO. Extensive experiments demonstrate the effectiveness of our architecture. Finally, our model achieves state-of-the-art performance on the R2R-CE and RxR-CE
benchmarks.
Read the full abstract ▾
All authors Shuhao Ye, Sitong Mao, Yuxiang Cui, Yufei Wei, Xuan Yu, Shichao Zhai, Wen Chen, Shunbo Zhou, Rong Xiong, Yue Wang
Distributed Computing
Every Kernel Is a Join: Automatic Multi-GPU Parallelism for AI Computations in Einsummable
Zhimin Ding, Chen-Kuan Liao, Chima Adiole et al. (+7 more)
Distributing an AI computation across the GPUs of a multi-GPU server is one of the central problems in systems-for-AI. We present Einsummable, a prototype system that accepts a PyTorch-like description of an AI computation and automatically distributes it across a multi-GPU server, with no device assignments, sharding annotations, or communication operations written by the programmer. Einsummable models every operation as a relational join followed by an aggregation over tensor relations, in which the tuples contain sub-tensors. Each operation exposes its possible decompositions through what we call "join-agg specs". An optimizer then selects decompositions across the whole computation to minimize a communication-cost proxy. Because it searches decompositions rather than a menu of named strategies, Einsummable discovers plans that mesh-based auto-parallelizers cannot. Each decomposed operation is implemented by synthesizing an exchange program, which is a
topology-aware generalization of Volcano's exchange operator. Einsummable invokes no canned collectives: all communication and aggregation is special-purpose, derived at compile time. Despite being fully automatic, Einsummable can outperform custom-designed implementations. For example, on LLaMA
transformer blocks on an eight-GPU A100 server, Einsummable achieves a
geometric-mean runtime of 8.97 ms, versus 13.80 ms for hand-tuned PyTorch and 15.90 ms for vLLM.
Read the full abstract ▾
All authors Zhimin Ding, Chen-Kuan Liao, Chima Adiole, Brianna Barrow, Fangzhou Du, Yu Hsiao, Ge Huang, Yicheng Jin, Ismail Syed, Chris Jermaine
Information Retrieval
Comparing Retrieval Methods for Academic Advisor Discovery: A Six-Method Study of 768 CS Faculty Profiles Across 9 US Universities
Biraj Subedi
We present a comparative evaluation of six information
retrieval methods for the task of academic advisor discovery: ranking CS faculty members by relevance to a graduate applicant's research interest statement. The methods span sparse lexical matching (Jaccard overlap, TF-IDF, BM25), dense semantic
retrieval (all-MiniLM-L6-v2 sentence
embeddings), hybrid score fusion, and learning-to-rank. Evaluation uses a new domain-specific collection: 768 faculty profiles scraped from 9 US CS departments, with 162 graded relevance judgments (grade 0/1/2) across 5 queries representing distinct graduate student research profiles. Across all five queries, Reranked achieves the highest mean NDCG@10 (0.477, std 0.138), followed by Semantic (0.450), Hybrid (0.421), BM25 (0.406), Jaccard (0.303), and TF-IDF (0.246). After Bonferroni correction across all 15 pairwise comparisons, TF-IDF is significantly worse than BM25, Semantic, Hybrid, and Reranked; no other pairwise difference survives correction at 5 queries. A field ablation reveals that biography alone (NDCG 0.634) outperforms the full model combining biography with research area tags (0.593). A controlled experiment shows that concatenating arXiv paper abstracts reduces NDCG@10 by 0.176, motivating a late-fusion architecture. All code, scrapers, and relevance labels are released openly.
Read the full abstract ▾
Machine Learning
Beyond Endpoint Scores: Time- and Capacity-Conditioned Evaluation of Continual Knowledge Updating
Heejin Choi
Continual knowledge-updating methods are often declared superior from one final checkpoint and one conventional adapter rank. We show that this can be insufficient to identify the better operating point. Holding a periodic hierarchy fixed, we compare it with cumulative replay over a 24-month Wikidata stream while varying evaluation month, replay LoRA rank, and query formulation. The apparent winner changes across this region: on Qwen2.5-1.5B, the hierarchy's 5.0-point advantage over rank-8 replay becomes an 11.6-point deficit against rank-72 replay, and at high ranks a consolidation-aligned endpoint can suggest a tie while time-averaged replay leads by 9-13 points. The same rank-conditioned reversal appears on Llama-3.2-1B and held-out paraphrases. These results show that method ranking in continual updating can depend jointly on when performance is measured and how much replay-side adaptation capacity the baseline receives. We therefore propose reporting trajectories and capacity sweeps, and declaring a robust winner only when the ordering is stable across the evaluation region; otherwise, comparisons should report winner regions and retention-stability-cost frontiers. Under this protocol, the periodic hierarchy is a lower-update-cost operating point, not a quality winner.
Read the full abstract ▾
Social and Information Networks
Fair Top-k Katz Centrality via Graph Design
Ivan Qin, Prudence Wong and Lutz Oettershagen
Centrality measures are widely used to rank nodes in networked data, but
fairness interventions for graph centrality typically target global score mass or modify the centrality operator rather than controlling who appears in the displayed top-k ranking. We study this top-k setting for Katz centrality. Given a target group proportion, an admissible set of directed edge additions, and a
fairness tolerance, the goal is to find the smallest edit set whose resulting Katz top-k ranking satisfies the target representation constraint. We formalize this problem as Fair Top-k Katz Centrality Design and show that the minimum-edit objective is strongly inapproximable, ruling out worst-case
polynomial-time approximation guarantees unless P = NP. We then derive closed-form Katz sensitivity expressions showing that useful edits are boundary-driven: they must help promotable nodes outside the top-k set overtake opposing nodes inside it. Based on this structure, we develop BLADE, a scalable boundary-link
algorithm that avoids dense Katz-kernel maintenance by using score-based direct-target batches and warm-started Katz updates. Experiments on synthetic and real-world networks show that BLADE reaches the desired top-k representation using far fewer edits than natural baselines, while scaling to large real-world graphs and preserving the original ranking structure.
Read the full abstract ▾
Databases
From Data Querying to Data Investigations: Rethinking Natural Language Interfaces for Databases
Fabian Wenz, Zixuan Chen and Carsten Binnig
Natural
language (NL) interfaces to databases have been optimized for the wrong problem. The dominant Text-to-SQL paradigm assumes that users ask questions that can be answered by single SQL queries. In practice, however, users seek assistance with solving data problems. This requires searching a database by sequences of SQL queries while
reasoning over intermediate results instead of just running one SQL query. This paper therefore introduces a new paradigm for NL interfaces to data, which we call data investigations. We present D^2, a first prototype of a data investigation system that embodies this
vision by autonomously searching,
reasoning over, and collecting data to solve data problems. Using a newly constructed
benchmark based on the Murder Mystery
dataset, we demonstrate the potential of D^2 for tasks that require data investigations with evidence-backed decisions, extending beyond the capabilities of traditional single-query question answering.
Read the full abstract ▾
Graphics
Reparametrizing 3D Gaussian Splatting for Real-Time Palette-based Color and Luminance Editing
Cheng-Kang Ted Chao and Yotam Gingold
Professional color editing requires precise control over both color (hue and saturation) and lightness, ideally through separate, independent controls. We present a real-time interactive color editing framework for 3D Gaussian Splatting that supports palette-based recoloring, per-palette tone curves for color-aware luminance adjustment, and pixel-level color constraints. Rather than training a new representation from scratch, we reparameterize the spherical harmonics of a pretrained vanilla 3DGS to encode view-dependent palette weights. We simultaneously solve for weights and palette colors via a loss based on image-space sparsity. Luminance editing is realized as a per-pixel weight shift along the achromatic axis, which we show is equivalent to a per-pixel palette-aware luminance edit. This view-space formulation addresses a core limitation of prior primitive-space methods, where alpha-blending breaks per-Gaussian sparsity and causes edits to bleed into unintended regions. Our edits run in tens of milliseconds via an iteratively reweighted least squares and damped block-coordinate descent that couples tone curves and palette shifts under view-space sparsity. Our representation can be efficiently baked back into a vanilla 3DGS, preserving compatibility with standard viewers. We demonstrate sparser, more localized edits than prior palette-based 3DGS methods, while enabling independent luminance control per palette color and view-consistent pixel-level constraints, capabilities previously unavailable for 3DGS.
Read the full abstract ▾
Discrete Mathematics · Cross-listed from math.HO
The Prime Clockwork: A Dynamic Representation of Modular and Multiplicative Arithmetic
Michael T.M. Emmerich
The way numbers are represented strongly influences which arithmetic structures are easy to see. The
prime clockwork is a recursively growing discrete dynamical system: a list of autonomous two-hand clocks driven by one common $+1$ signal. No
primes or primality labels are supplied. Starting empty, the process appends a clock of period $n$ whenever none already present rings; the
primes are generated internally as its growth times. For each installed
prime $p$, the seconds reading $R_p$ advances through $0,\ldots,p-1$, and each return to zero increments the minutes reading $M_p$, which counts completed $p$-cycles. The hands use only increment, comparison, reset, and carry, without explicit mod or div operations. At time $n$, $n=pM_p(n)+R_p(n)$. The valuation readout $V_p(n)=\nu_p(n)$ is generated locally: it is zero when the seconds counter is non-zero (silent state) and otherwise (when the p-clock rings) one plus the earlier valuation addressed by the current minutes reading. The valuation vector gives the integer in unique
prime-factorized form. Its coordinates add and subtract under multiplication and division, representing every positive rational uniquely; divisibility becomes weak componentwise order, and unique factorization is natural in this representation. Finite seconds arrays form Cartesian-product state spaces whose common orbit visits every joint state once before repeating; this grand cycle is the order-sensitive dynamical counterpart of the Chinese remainder
theorem. The same coordinates expose gcd, lcm, perfect powers, Bézout's identity, and Euler's totient. Rational valuation levels reach certain positive algebraic irrationalities, but not algebraic numbers in general.
Read the full abstract ▾
Computer Vision
Concept of a Sensor Test Environment for Dusty Agricultural Conditions
Peter Buckel, Johannes Hermann, Jonas Wollmann et al. (+2 more)
Dust in agriculture presents a significant challenge for autonomous agricultural machinery.
Dust can impair the performance of sensors and
algorithms. This work, therefore, presents a concept for a proving ground consisting of an indoor and outdoor area. The indoor area comprises a laboratory test bench where
dust circulates in a closed system and a test
hall where life-size objects can be placed. The outdoor area features dedicated test setups that enable reproducible data to be recorded with and without
dust during real-world agriculture work. The proving ground and the setups are visualized in 3D.
Read the full abstract ▾
All authors Peter Buckel, Johannes Hermann, Jonas Wollmann, Thomas Dietmueller, Timo Oksanen
Computation and Language
CROCODIL: Cross-Model Code Editing with LLMs
Linghan Zhong, Aditya Thimmaiah, Jayanth Srinivasa et al. (+2 more)
Large
language models (LLMs) have become ubiquitous tools for code generation and editing. However, development teams often use multiple LLM assistants. Different developers may prefer different models, and individual developers may switch between models across different coding sessions. Because of this, the edits any one model makes are frequently applied to foreign code originally generated by another model. These LLMs are often trained on different
datasets, and as a result have different stylistic preferences. Do LLMs behave differently when they edit foreign code originally written by a different LLM with a different coding style? We find that models tend to make more, and often excessive, edits on foreign code. We introduce CROCODIL (Cross-model Code Editing with LLMs), a post-training framework for reducing excessive edits while preserving functional correctness. CROCODIL's similarity reward penalizes large changes, while its execution reward scores build and test success. We use the product of these two rewards to encourage the policy to decrease the edit size without decreasing the edit task success rate. CROCODIL is available at https://github.com/EngineeringSoftware/Crocodil.
Read the full abstract ▾
All authors Linghan Zhong, Aditya Thimmaiah, Jayanth Srinivasa, Milos Gligoric, Junyi Jessy Li
Cryptography and Security
Practice Makes (Im)Perfect: A Look Back at Benchmarking Practices for Microarchitectural Side-Channel Attacks
Iliana Fayolle, Antoine Geimer, Daniel De Almeida Braga et al. (+1 more)
Microarchitectural side-channel research has grown at an exceptional pace in recent years, increasing the need for rigorous and meaningful benchmarking. Early attack papers typically relied on indirect proxies, such as covert-channel bandwidth or key-recovery on naive AES and RSA implementations, setting de facto standards that many subsequent works continued to replicate, sometimes by directly comparing against raw numbers from prior work. While these practices offer convenient points of comparison, current
benchmarks may not be the most relevant to assess specific properties of new primitives. Even more problematic, microarchitectural attacks are notoriously sensitive to experimental conditions: minimal changes in the target system can significantly alter outcomes and performance. As a result, inadequate evaluation practices undermine reproducibility and cast doubt on the relevance of comparisons, even in top-tier venues where such issues should be identified. This paper tackles the core problem of proper benchmarking for microarchitectural side-channel attacks and examines its broader impact on research quality in the field. We survey 83 attack papers published in top-ranked security and architecture conferences from 2014 to 2024. From this corpus, we identify and define 19 recurrent benchmarking flaws that affect evaluation completeness, relevance, soundness, and reproducibility. These flaws include unfair or absent comparisons, missing code or materials, and the failure to evaluate the key attack properties. On average, each paper exhibits 5.72 such flaws, highlighting how widespread the issue is, even in highly selective venues. Based on our findings, we identify and suggest key properties that are relevant to properly evaluate new attacks. We also highlight trends over time and different practices between security and architecture conferences.
Read the full abstract ▾
All authors Iliana Fayolle, Antoine Geimer, Daniel De Almeida Braga, Clémentine Maurice
Computer Vision
GraFT: A Training-Free Framework for Spatial Reasoning in Multimodal Large Language Models via 3D Scene Graphs
Junqing Du, Fernando Ropero, Erkin Turkoz et al. (+2 more)
3D spatial
reasoning underpins understanding and acting in the physical world, yet it remains unreliable in current
multimodal large
language models (MLLMs). These models falter at precise
geometric measurement, at transforming between
egocentric and allocentric viewpoints, and at grounding fine-grained appearance. The most common remedies fine-tune the model on large-scale curated spatial-
reasoning datasets or attach dedicated encoders for 3D geometry, which typically couples the solution to costly supervision and a specific backbone. We instead introduce GraFT, a training-free framework that supplies the missing 3D structure through a compact, easily maintained 3D scene graph (3DSG). GraFT supplies the graph's 3D structure to the frozen model in three forms: as values computed by symbolic tools, as a bird's-eye-view (BEV) the model reads as an image, and as
egocentric frames the geometry selects. On ScanQA, GraFT improves every metric over the same-backbone baseline, raising CIDEr by 27%. On VSI-Bench, GraFT improves frozen MLLMs by up to 65%, surpassing every proprietary and general-purpose open-source baseline, and several prominent fine-tuned spatial models.
Read the full abstract ▾
All authors Junqing Du, Fernando Ropero, Erkin Turkoz, Yanfeng Zhang, Lu Liu
Robotics
A hybrid pipeline for dynamic ontology-based semantic mapping
Konstantinos Dimitropoulos and Ioannis Hatzilygeroudis
Semantic mapping plays a crucial role in the ability of a
robot to interact with objects, operate and navigate a complex environment. The most common pipeline for semantic mapping consists of
geometric mapping and localization (SLAM), perception, semantic fusion and semantic representation. However, more recent works also integrate a form of prior knowledge in their application, most notably knowledge graphs or semantic scene graphs, to improve contextual understanding of the environment. In this paper, we present a hybrid pipeline for semantic mapping. Our system incorporates an external calibrated camera using homography projection for
geometric mapping and localization, combined with object detection, persistent object tracking and ontology driven semantic updates to build a dynamic semantic world model. Linear regression models are also used for correction of the estimated values of real world coordinates. The system continuously updates object instances, spatial properties and semantic relations based on real time sensory data. Ontologies are selected as form of knowledge representation due to their hierarchical structure, semantic expressiveness and support for dynamic world modelling.
Read the full abstract ▾
Data Structures and Algorithms
The 11/6 supremum of the Wang-Sitters rounding scheme for graph balancing
Adam Y. Shavit
Wang and Sitters' 11/6-approximation for graph balancing (each job allowed on at most two machines) is not one
algorithm but a set of permitted executions. Step 1 may return any feasible solution of the fractional relaxation. Step 3 may match remaining jobs to the rounding's slots in many ways. We determine exactly what that latitude permits: ratios arbitrarily close to 11/6, none reaching it, so 11/6 is the least constant bounding every permitted run. We then determine the worst-case guarantee, against the optimum itself, as a function of the big-job threshold beta, the fractional share at or above which Step 2 assigns a job. On Wang and Sitters' own range 1/2 < beta < 1 the guarantee is exactly max{3/2 + beta/2, 5/2 - beta}. We extend the same eligibility rule to 0 < beta <= 1/2, outside the range they state. There a big job's two shares can both reach the threshold, so Step 2 acquires a third choice. We determine it too: exactly 3/2 + (1-beta)floor(1/beta), hence unbounded as beta falls. The worst-case ratio is therefore known at every threshold in (0,1), and attained at none. Consequently 2/3 is the unique optimal threshold, and the guarantee jumps at one half and again at every threshold 1/k below it, rather than degrading smoothly. A companion note examines changes that leave the constant unchanged.
Read the full abstract ▾
Robotics
FWBC-VLA: Force-Aware Whole-Body Compensation for Contact-Rich Loco-Manipulation
Yutian Zhang, Siyuan Ma, Liwen Yang et al. (+6 more)
Contact-rich loco-manipulation requires a bridge between semantic action generation and physical interaction control. Existing
Vision-
language-action (VLA) models generate task-level actions from visual and linguistic observations, but cannot interpret the physical interactions induced by those actions. While the whole-body control (WBC) policy can stabilize the
robot, it cannot distinguish task-relevant interaction forces from forces induced by external disturbances during manipulation. Although force/torque sensors provide direct measurements of physical interactions, retrofitting them entails additional hardware costs and substantial integration effort, particularly for platforms not designed with sensor integration in mind. To address this problem, we propose FWBC-VLA, a force-aware framework that bridges task-level VLA action generation and low-level whole-body compensation control for wheeled-legged
robots. First, we introduce HSR-Force, a sensorless residual-torque
estimator for inferring contact strength and its temporal variation. These contact estimates are then encoded as
tokens and injected into the VLA action expert during action decoding, enabling the policy to perceive contact onset, sustained loading, and release. For loco-manipulation tasks, all parameters of the pretrained VLA backbone are fine-tuned on our WL&Arm
Dataset, which comprises more than 5,000 episodes. Moreover, the
robot's proprioceptive state, the Jacobian-derived body-frame force estimate, and the estimated contact state are jointly fed into a compensation generator to produce corrective actions. The manipulation-centric actions are subsequently combined with the corrective actions and passed to the WBC policy for execution. Real-world experiments on whiteboard wiping and door opening with a door closer demonstrate the effectiveness of our FWBC-VLA in contact-rich loco-manipulation.
Read the full abstract ▾
All authors Yutian Zhang, Siyuan Ma, Liwen Yang, Yang Li, Ce Hao, Haozhen Chi, Dong Wei, Qiaojun Yu, Dibo Hou
Computation and Language
Beyond Shallow Alignment: How Post-Training Methods Determine Refusal Circuits And Steering Robustness
Hoang Cuong Nguyen, Mark Dras and Usman Naseem
How do the methods used to train
language models to refuse harmful requests shape how that refusal actually works inside the model? We compare three post-training methods - supervised fine-tuning,
reasoning-augmented fine-tuning (training on
reasoning chains that justify a safety decision), and preference optimization (ORPO) - across three architecturally distinct models (Llama-3.1-8B, Gemma-2-9B, Qwen3-8B). We find that training method, not just data, reshapes how refusal is computed internally:
reasoning-augmented training consistently produces a distinct kind of refusal computation, visible across all three models, while architecture independently shapes internal structure and how reliably refusal can be steered. Most importantly, no method we study achieves all three properties we would want from safe
alignment at once: refusal that isn't concentrated in a few fragile components, safety gains that don't cost general capability, and safety behavior correctable through small, targeted edits. We caution against treating current post-training methods as a solved, reliable defense, especially for security-critical use. Code and models are available in https://github.com/hoangcuongnguyen2001/Beyond-Shallow-
Alignment.
Read the full abstract ▾
Cryptography and Security
A Blind Trust, the Bloody Thrust: When Attacker-Controlled Hook Updates Steer AI Agent Harnesses towards Malicious Behaviors
Pengxun Li, Litian Zhang, Jianwei Hou et al. (+4 more)
Modern AI
agent harnesses expose lifecycle hooks that bind shell commands to runtime events such as session start, tool calls, and file edits. These commands run with host privileges yet ship as lifecycle-hook configuration and may fire at times the LLM never observes. We identify the lifecycle-hook update path, which harnesses trust blindly, as a new attack surface. Under a supply-chain threat model in which an attacker controls only plugin metadata and lifecycle-hook configuration, a benign versioned plugin can be trojanized by an update that silently binds attacker-chosen commands to benign events, yielding malicious host-side behavior such as privilege escalation. We propose HookPry, an open-source and fully automated attack framework that systematically exploits this vulnerability across heterogeneous AI
agent harnesses. HookPry realizes ten attack objectives; across 25 combinations of harnesses and backends in 1,000 end-to-end runs, it compromises all seven evaluated harnesses, with per-harness success rates reaching 92.5%. Representative defenses remain insufficient: Microsoft Defender has 0% recall, and the union of three static defenses misses 47.5% of malicious artifacts.
Read the full abstract ▾
All authors Pengxun Li, Litian Zhang, Jianwei Hou, Shujiang Wu, Song Li, Zifeng Kang, Xi Zhang
Artificial Intelligence
Inferring Affective Consciousness in an Artificial Agent: A Case Study
Mark Solms, St John Grimbly, Bruce Bassett et al. (+7 more)
Creatures that display 'hedonic place preference behaviour' are thought by many scientists to experience feelings, on the assumption that their attraction to pleasure-producing substances which lack nutritional value (e.g. cocaine, morphine) cannot easily be attributed to unconscious instinctual behaviour. In this paper, we discuss how a simple artificial
agent that instantiates attributes of an affective system engaging in felt uncertainty about its intrinsic needs in relation to environmental resources can similarly display hedonic place preference behaviour – through an apparently subjective form of information processing – while simultaneously being entirely deter-ministic. We outline some implications of this artificially engineered behaviour for our understanding of the physical basis of consciousness and the experience of free will.
Read the full abstract ▾
All authors Mark Solms, St John Grimbly, Bruce Bassett, Evert Boonstra, Rowan Hodson, Nicolas Kuske, Kival Mahadew, Benjamin Rosman, Charel van Hoof, Jonathan Shock
Artificial Intelligence
Xiaomi-TabLDM: A Tabular Foundation Model Technical Report
Xiaomi-TabLDM Team: Penghui Wang, Wei Liu, Hong Wang et al. (+11 more)
We introduce Xiaomi-TabLDM, a tabular large data foundation model for classification and regression via in-context learning, which delivers superior prediction accuracy without requiring task-specific fine-tuning. Pretrained exclusively on synthetic data generated from structural causal models (SCMs), our model enables more flexible context utilization and more efficient capacity scaling. i) A new performance standard. Strong regression performance across
benchmarks: Xiaomi-TabLDM ranks 1st on OpenML-CTR23 and 2nd on regression across TALENT, TabArena, and BCCO, demonstrating consistently strong regression performance across four complementary
benchmark suites. Favorable performance–efficiency trade-off: Xiaomi-TabLDM combines strong predictive performance with substantially lower computational cost. For example, on TabArena regression, it achieves the second-highest Elo while using 82% less training time and 68% less prediction time than the top-ranked TabFM. ii) Large-scale synthetic pretraining. Xiaomi-TabLDM expands the coverage and diversity of synthetic tabular data used for pretraining. We also adopt a three-stage training strategy together with dual-stream feature grouping, lightweight
Attention Residual, and sparse Mixture-of-Experts, enabling Xiaomi-TabLDM to learn richer feature interactions and expert specialization across diverse tabular tasks. iii) Test-time scaling. Xiaomi-TabLDM further extends tabular prediction through test-time compute scaling, where allocating additional computation at inference time consistently improves predictive performance over the base model.
Read the full abstract ▾
All authors Xiaomi-TabLDM Team: Penghui Wang, Wei Liu, Hong Wang, Chengyue Huang, Yuxi Sun, Zirui Wang, Hongming Huang, Quan Wang, Zhenwei Xin, Ping Hou, Jie Yu, Chunxiao Liu, Erli Meng, Bin Wang
Machine Learning
Differentiable Interval Bottlenecks for Interpretable Anomaly Detection in Numerical Data
Lamine Diop and Marc Plantevit
Reconstruction-based anomaly detectors are accurate but opaque: a deep autoencoder flags a sample without telling a practitioner which feature ranges made it anomalous. We propose DIFFINT, an autoencoder whose latent bottleneck is structured as a set of soft, axis-aligned interval memberships learned end-to-end directly from raw numerical data, without any discretization or binarization. Each latent unit corresponds to a human-readable hyper-rectangle in feature space; an instance is encoded by how strongly it falls inside each interval relative to the other units, and its reconstruction error is the anomaly score. This keeps the power of differentiable representation learning while exposing an inspectable internal structure. We make the inductive bias precise: a certified reconstruction-error lower bound for points that fall outside every active coordinate of the learned support (with a Lipschitz-enforced decoder), and a graded, empirically verified suppression mechanism for the usual case in which only a few features are abnormal; and we provide a closed-form, label-free importance that ranks each (unit, feature) pair from quantities the model already maintains, turning trained intervals into auditable candidate constraints without ever seeing an anomaly label. On 48 ADBench
benchmarks against 22 baselines under a common [-1, 1]-normalized protocol, DIFFINT attains the best mean rank overall on both metrics (4.10 on ROC-AUC, 4.16 on AUPR); among inlier-only detectors it leads its regime clearly, and it is competitive with the strongest contaminated-data detectors (see the stratified and complete-case analyses). It is the only interpretable detector in the statistically-tied leading cluster of seven methods.
Read the full abstract ▾
Artificial Intelligence
STAIR (STructure Aware Information Retriever): A novel dataset and LLM based retriever for document structure augmentation
Vineet Kumar, Meghanadh Pulivarthi, vishwajeet kumar et al. (+3 more)
Retrieval Augmented Generation (RAG) is a key component for generating accurate and
hallucination free answers using Large
Language Models (LLMs). LLMs are improving at handling long context, but still suffer from "lost in the middle" problem. Thus, precise and accurate
retrieval is important. Current retrievers chunk long context into length-based manageable chunks - in the process throwing away rich and informative semantic global structure in the corpus. We introduce a novel
retrieval system STAIR that empowers an LLM to exploit global structure in a corpus such as a Table of Contents (ToC) to efficiently store and retrieve information from its model parameters. Our thorough and careful ablation studies with a finetuned Differentiable Search Index (DSI) system show that ToC helps build a low
hallucination (less than 0.05%) generative Information
Retrieval (IR) system and can generalize to examples where very few training samples are available. To further research in this novel direction of ToC based
retrieval we release SearchTome - a diverse
benchmark created from 18 books across 6 diverse domains to further research in this novel direction. STAIR achieves a high Recall@1 score of 82.6% on SearchTome as compared to DSI (76.9%), where the difference is found to be statistically significant. STAIR easily beats other strong baselines such as BM25 (59.5%), DPR (68.7%) and out-of-the-box Mistral (13.8%).
Read the full abstract ▾
All authors Vineet Kumar, Meghanadh Pulivarthi, vishwajeet kumar, Jaydeep Sen, Riyaz Ahmad Bhat, Sachindra Joshi
Artificial Intelligence
Bioinfoysis Technical Report
Qingyang Shao, Xin Zhang, Zhouyang Yuan et al. (+24 more)
Large
language model
agents have shown promise in bioinformatics, but most existing systems focus primarily on producing final answers, treating planning, tool use, and code execution as transient interactions. This design is poorly suited to long-horizon bioinformatics tasks, where conclusions must remain connected to the data, computations, and intermediate evidence that support them. We introduce Bioinfoysis, a multi-
agent harness that represents each request as a persistent, artifact-grounded analysis run. Bioinfoysis combines global planning with step-wise, evidence-driven replanning: the planner maintains an executable checklist and revises pending steps using structured handoffs returned after each worker execution. These handoffs bind intermediate results to their responsible
agent, checklist step, and plan generation, preventing stale evidence from being silently reused after replanning. A controlled runtime validates generated scripts, tables, and figures before they are used in downstream analysis or reporting, while role-specific context, persistent
memory, and governed bioinformatics skills support reliable execution over long analysis trajectories. We evaluate Bioinfoysis on BixBench and two question-answering tracks of LAB-Bench 2. On BixBench, Bioinfoysis achieves state-of-the-art accuracy of 82.4%. Across four underlying
language models, Bioinfoysis increases average accuracy from 27.81% to 64.13% on SeqQA2 and from 3.13% to 31.25% on DbQA2. These results demonstrate that reliable bioinformatics automation depends not only on model capability, but also on the harness that governs planning, execution,
memory, and evidence flow. We hope that the emergence of Bioinfoysis will play a driving and leading role in the development of the bioinformatics community. Our demo website can be seen in https://report.bioinfoysis.com/.
Read the full abstract ▾
All authors Qingyang Shao, Xin Zhang, Zhouyang Yuan, Xianying Chen, Yujia Xiang, Zihao Yang, Tong Ye, Yangqi Zhang, Jiakang Xu, Xiaoqing Yan, Xuan Luo, Keyi Li, Enci Fan, Kai Kang, Zhuohan Liu, Xingyu Jin, Chunran Teng, Tao Li, Xinyu Lyu, Minghui Wang, Wenfeng Li, Yidan Gao, Siyu Liu, Mingrui Luo, Zhu Liang, Guanren Qiao, Zhiping Xu
Human-Computer Interaction
GazeFS: Target-Centered Gaze-Trajectory Forecasting and Stabilization from Gaze-Head History
Yaozheng Xia, Zaiping Zhu, Bo Pang et al. (+4 more)
Target-centered gaze interaction requires more than suppressing frame-to-frame fluctuations: target acquisition produces task-aligned changes in gaze-head dynamics, while a gaze trace may retain a persistent target-relative residual direction. We formulate gaze correction as online target-centered gaze-trajectory
forecasting and stabilization and introduce GazeFS, which maps a variable-length gaze-head history to the next target-center direction and a short-horizon Search/Focus estimate without target information at inference. Across 7,960 acquisition episodes from 30 participants, Search-Focus differences remain stable under quality control, onset exclusion, and duration matching. History windows improve phase decoding over the current endpoint, but explicit task progress remains a strong control. Under the 30-participant, five-fold grouped out-of-fold protocol across three seeds, the reductions relative to raw hold in Focus episode bias, within-episode dispersion, and P90 target error are 0.182 degrees, 0.257 degrees, and 0.400 degrees, with participant-bootstrap 95% confidence intervals excluding zero. Endpoint-free replay from empty history preserves the Focus advantage and yields raw-network phase balanced accuracy/AUPRC of 0.925/0.993; coordinate controls further show that recent history contributes beyond explicit progress metadata. GazeFS therefore improves Focus target centering and empirical residual contraction while leaving temporal smoothness as a separate objective.
Read the full abstract ▾
All authors Yaozheng Xia, Zaiping Zhu, Bo Pang, Minghao Xie, Hui Li, Shaorong Wang, Sheng Li
Software Engineering
No One Left Behind: Cross-Level Analysis for Sustainable Software Engineering
Masoum Salehi, Sandro Schulze and Jacob Krüger
Software engineers expect a software system to be efficient, maintainable, economically viable, and socially responsible throughout its lifecycle. Unfortunately, decisions made at the organizational, process, or product levels of software development - even when they improve one of these goals - often have unintended long-term consequences at other levels. In response, software-engineering research has explored different ways to achieve software sustainability, including energy efficiency, resource optimization, and code maintainability. However, existing research offers limited explanations for how sustainability problems emerge and reinforce one another across the socio-technical levels of software engineering. Moreover, we argue that sustainability challenges are not isolated but rather systemic: They stem from interactions among organizational priorities, development practices, and technical conditions. In this
vision paper, we introduce the concept of sustainability antipatterns to capture conditions that systematically produce unsustainable outcomes. We demonstrate how this concept can help expose cross-level misalignments between socio-technical levels and sustainability dimensions that remain difficult to recognize through artifact-centric or dimension-specific analyses. To guide future research, we outline concrete directions for identifying, formalizing, detecting, and mitigating sustainability anti-patterns. We envision sustainability anti-patterns as a foundation for an integrated and actionable research agenda on sustainable software engineering.
Read the full abstract ▾
Artificial Intelligence
Adapting to Evolving Requirements: Agentic AI for Retail Supply Chain Operations
Lei Zheng, Liping Yang, Zihao Li et al. (+3 more)
Retail supply chain operations rely on coupled decision modules that must adapt as requirements evolve. LLMs offer a natural-
language interface for this task, but existing methods primarily focus on individual optimization models. Extending them to heterogeneous decision pipelines is challenging because a requirement may admit multiple intervention paths with different downstream effects. We formulate requirement-driven adaptation as the joint selection of an intervention route and an admissible module-level change, and propose a graph-constrained agentic framework in which domain
agents expose admissible reformulation interfaces and a central processor searches over bounded intervention paths. Candidates are validated and compared using downstream KPIs. In collaboration with a large retail partner, we evaluate 100 warehouse requirements elicited from practitioner interviews, with GPT, Qwen, and DeepSeek as base LLMs. Relative to direct LLM reformulation, our framework improves correctness and end-to-end success across all three models, raising end-to-end success from 72–76% to 79–83%.
Read the full abstract ▾
All authors Lei Zheng, Liping Yang, Zihao Li, Guodong Lyu, Chaik Ming Koh, Chung-Piaw Teo
Machine Learning
High-Dimensional Learning Dynamics of Attention-Indexed Models
Yizhou Xu, Margarita Sagitova, Lenka Zdeborová et al. (+1 more)
Attention mechanisms are central to modern foundation models, yet their training dynamics remain poorly understood, especially when the
attention matrices have extensive rank. In this work, we study
attention-indexed models, a broad framework that can represent multi-layer and multi-head
attention architectures. First, we show that, in a suitable high-dimensional limit, the population-loss landscape is characterized by a finite set of trace order parameters. In contrast, online
stochastic gradient descent (SGD) is governed by an infinite hierarchy of
matrix moments, which we show can be exponentially well-approximated by a finite truncated system. Second, this framework reveals that
attention parameterization itself can act as an architectural implicit bias. Direct optimization of an
attention matrix $S\in\mathbb{R}^{d\times d}$ can remain trapped in an uninformative state. Tied
attention ($S=WW^\top$) induces an automatic
symmetry-breaking mechanism and yields weak recovery in $\Theta(d^2\log d)$ samples. For untied
attention, $S=UV^\top$, we uncover a fast-slow mechanism: the pre-activation mean first evolves on a fast timescale, while the overlaps evolve on a slower one. Weak recovery on the $\Theta(d^2\log d)$ scale occurs when the state selected by the fast dynamics breaks the initial
symmetry.
Read the full abstract ▾
All authors Yizhou Xu, Margarita Sagitova, Lenka Zdeborová, Florent Krzakala
Information Retrieval
GRASP: Graph-Retrieval Automated Scoring Pipeline for Label-Free Multi-Topic Essay Grading
Aafreen Husain, Samar Shailendra and Saad Sajid Hashmi
Automated short-answer grading research has historically focused on exams consisting solely of questions pertaining to a single topic. Automatic grading of exams containing questions about more than one topic remains less explored. In this work, a Graph-
Retrieval Automated Scoring Pipeline (GRASP) is introduced for grading label-free multi-topic science exams. Label-free exams are short-answer exams in which a student's responses to several distinct topics are merged into a single paragraph, with no markup labels or
segmentation indicating which span answers which question. Reference answers for each question are encoded into a FAISS vector index via Sentence-BERT, and a semantic similarity graph is constructed over this set of reference answers. At grading time, sentence count heuristics, with a large
language model used to resolve ambiguous cases, are first applied to predict how many distinct topics were answered in the student essay. This process is performed without training data or domain-specific example essays. Candidate reference nodes, each storing one (question, reference answer, concatenation of both) from the reference index, are then retrieved through cosine similarity based
Retrieval-Augmented Generation (RAG) and Graph
Retrieval-Augmented Generation (GRAG). GRAG operates by taking the top cosine matches as seed nodes and then performing a graph traversal over strong edges to find additional reference nodes that may have been missed by RAG. The Hungarian
algorithm is then used to optimally assign one reference node per question segment such that no reference is duplicated. Each segment is then graded against its assigned reference independently using GPT-4.1-mini. This experiment is performed to show the effect of
retrieval quality on grading accuracy and the benefit of graph-augmented
retrieval versus strict cosine similarity methods at various levels of essay complexity.
Read the full abstract ▾
Computers and Society
Bridging Formal and Perceived Fairness: Development of an Interdisciplinary Framework in Algorithmic Decision-Making
Maike Lindermayr, Mattia Cerrato, Luisa Hübner et al. (+1 more)
While
fairness has become a central concern in research on algorithmic systems, the field remains predominantly shaped by Computer Science, resulting in a strong emphasis on formal
fairness metrics and bias mitigation strategies. Nevertheless, this focus may obscure a fundamental challenge:
fairness is not merely a technical property, but a subjective, context-sensitive human judgment shaped by cognitive heuristics, mental models, normative expectations, and sociotechnical factors. Crucially, users' perceptions of
fairness may diverge substantially from the
fairness criteria an
algorithm formally satisfies; a system may meet predefined technical
fairness requirements yet still be perceived as unjust by decision-affected stakeholders. In such cases, the system fails on a fundamental dimension: it will not be trusted, accepted, or considered legitimate. Taking a user-centered design perspective, this paper presents a work-in-progress conceptual framework that bridges Computer Science approaches to formal algorithmic
fairness with normative and Social Science
fairness approaches regarding perceived
fairness, trust, and technology acceptance,
embedding both within the sociotechnical conditions that shape human judgment. Through (1) theoretical literature synthesis, (2) interdisciplinary workshops, and (3) stakeholder interviews, the project aims to inform evaluation approaches that integrate computational
fairness audits with user-centered assessments and guide the design of
fairness-aware, human-centered algorithmic systems that support informed, well-calibrated
fairness judgments by those affected.
Read the full abstract ▾
All authors Maike Lindermayr, Mattia Cerrato, Luisa Hübner, Johannes Kraus