Artificial Intelligence · Cross-listed from cs.CV
SARLO-80: Worldwide Slant SAR Language Optic Dataset 80cm
Solène Debuysère, Nicolas Trouvé, Nathan Letheule et al. (+2 more)
Multimodal foundation models have advanced rapidly thanks to large optical
benchmarks, but comparable resources for synthetic aperture radar (SAR) remain limited. Existing SAR–optical
datasets largely rely on low-resolution, intensity-only Ground Range Detected~(GRD) products and do not preserve complex-valued SAR measurements or native acquisition geometry, which restricts physically grounded
multimodal learning. In particular, large-scale public
datasets combining very-high-resolution (VHR) SAR SLC, aligned optical imagery, and natural-
language descriptions are still lacking. We present a VHR SAR–optical–text
dataset built from open-access Umbra spotlight acquisitions distributed as Sensor Independent Complex Data (SICD). From around 2,500 worldwide scenes (VV/HH, 20cm–2m native resolution), we standardize all SAR data to an 80cm slant-range grid via band-limited FFT resampling and tile the imagery into 1024 by 1024 patches. For each SAR patch, we retrieve a high-resolution optical tile and warp it into the SAR grid using local coordinate correspondences for local pixel-level
alignment. We further generate three caption variants (SHORT/MID/LONG) per sample to support
vision–
language training and evaluation. Our
dataset contains 119,566 triplets (complex and amplitude slant-range SAR patch, aligned optical patch, natural-
language description) covering 257 locations across 72 countries and a broad range of land types and infrastructures. We release fixed train/validation/test splits and the full preprocessing and baseline code to enable reproducible
benchmarks for
multimodal alignment on cross-modal
retrieval and conditional generation in native SAR geometry. The
dataset is publicly available on the Hugging Face Hub at https://huggingface.co/
datasets/ONERA/SARLO-80.
Read the full abstract ▾
All authors Solène Debuysère, Nicolas Trouvé, Nathan Letheule, Elise Colin, Georgia Channing
Artificial Intelligence · Cross-listed from cs.CR
Sovereign Execution Broker: Enforcing Certificate-Bound Authority in Agentic Control Planes
Jun He and Deying Yu
Autonomous
agents are increasingly connected to cloud, deployment, and data-control workflows, but production mutation authority should not reside inside non-deterministic
reasoning processes. Existing access-control mechanisms authorize identities, while assurance layers certify proposed actions; neither alone provides a mandatory enforcement point for certified authority at the moment of mutation. This paper introduces the Sovereign Execution Broker (SEB), a runtime enforcement boundary for certificate-bound agentic infrastructure. SEB consumes certificates issued by the Sovereign Assurance Boundary (SAB), verifies that the requested mutation matches the certified execution contract, checks validity windows, policy epochs, revocation epochs, and live-state drift, mints scoped execution identity, invokes infrastructure APIs, and records signed decision and outcome records. By separating proposal, admission, and execution, SEB turns certified authority into a short-lived, revocable, auditable runtime capability, provided that production mutation APIs reject non-broker identities. We present the SEB execution model, certificate and replay-verification predicates, scoped identity semantics, bypass-prevention deployment patterns, failure behavior, and a concrete prototype implementation. We evaluate the prototype on AWS and Kubernetes clusters, measuring latency overheads, revocation propagation, drift detection, and security under fault injection.
Read the full abstract ▾
Artificial Intelligence
FlowEdit: Associative Memory for Lifelong Pronunciation Adaptation in Flow-Matching TTS
Harshit Singh, Ayush Pratap Singh and Nityanand Mathur
Flow-matching text-to-
speech systems achieve remarkable zero-shot quality but remain static after deployment: pronunciation errors on out-of-vocabulary proper nouns persist unless the model is retrained. We introduce FlowEdit, a life-long adaptation framework for frozen flow-matching TTS that learns pronunciation corrections as latent conditioning edits rather than weight updates. When corrective feedback is provided, FlowEdit optimizes a
token-level perturbation in the text
embedding space, then stores the correction in a Modern Hopfield Network serving as content-addressable episodic
memory. At inference, corrections are retrieved via soft
attention with a similarity gate, enabling fuzzy morphological matching. On our curated
benchmark of 312
multilingual proper nouns across 18
language families, FlowEdit reduces target-word Phoneme Error Rate by 92.7% relative to the zero-shot baseline while maintaining identical general-
speech quality. Corrections complete in approximately 15 seconds on a single GPU.
Read the full abstract ▾
Artificial Intelligence
Multi-LCB: Extending LiveCodeBench to Multiple Programming Languages
Maria Ivanova, Pavel Zadorozhny, Rodion Levichev et al. (+5 more)
LiveCodeBench (LCB) has recently become a widely adopted
benchmark for evaluating large
language models (LLMs) on code-generation tasks. By curating competitive programming problems, constantly adding fresh problems to the set, and filtering them by release dates, LCB provides contamination-aware evaluation and offers a holistic view of coding capability. However, LCB remains restricted to Python, leaving open the question of whether LLMs can generalize across the diverse programming
languages required in real-world software engineering. We introduce Multi-LCB, a
benchmark for evaluating LLMs across twelve programming
languages, including Python. Multi-LCB transforms Python tasks from the LCB
dataset into equivalent tasks in other
languages while preserving LCB's contamination controls and evaluation protocol. Because it is fully compatible with the original LCB format, Multi-LCB will automatically track future LCB updates, enabling systematic assessment of cross-
language code generation competence and requiring models to sustain performance well beyond Python. We evaluated 24 LLMs for instruction and
reasoning on Multi-LCB, uncovering evidence of Python overfitting,
language-specific contamination, and substantial disparities in
multilingual performance. Our results establish Multi-LCB as a rigorous new
benchmark for multi-programming-
language code evaluation, directly addressing LCB's primary limitation and exposing critical gaps in current LLM capabilities.
Read the full abstract ▾
All authors Maria Ivanova, Pavel Zadorozhny, Rodion Levichev, Ivan Petrov, Adamenko Pavel, Ivan Lopatin, Alexey Kutalev, Dmitrii Babaev
Artificial Intelligence · Cross-listed from cs.CR
Efficient and Sound Probabilistic Verification for AI Agents
Alaia Solko-Breslin, Pramod Kaushik Mudrakarta, Mihai Christodorescu et al. (+2 more)
Securing AI
agents that operate in complex digital environments has become a critical need, and runtime monitoring approaches that formulate and enforce policies expressed in a formal
language like Datalog offer a promising solution. However, existing approaches are restricted to deterministic policies. In many practical applications of AI
agents, there is a need to enforce security policies in the face of ambiguity, leading to probabilistic predicates or state transitions (for example, a declassifier or Personally Identifiable Information (PII) detector that has some failure probability on each invocation). Furthermore, in many such applications, one cannot easily make the independence assumptions necessary to invoke prior work on probabilistic inference in Datalog. We address this by introducing a sound and efficient framework for such verification based on distributionally robust optimization, computing sound upper bounds on the probability of policy violation regardless of possible correlations between predicates. On standard
benchmarks for terminal and tool calling
agents, we demonstrate that our approach outperforms prior art and improves the security-utility trade-off while ensuring rigorous bounds on the probability of policy violation.
Read the full abstract ▾
All authors Alaia Solko-Breslin, Pramod Kaushik Mudrakarta, Mihai Christodorescu, Somesh Jha, Krishnamurthy Dj Dvijotham
Artificial Intelligence
What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?
Sihui Dai and Mann Patel
Prior work has shown that in-context demonstrations can jailbreak
language models, but it remains unclear how models interpret different types of compliance demonstrations. We study this by mixing benign compliance demonstrations (non-harmful request, helpful response) with harmful compliance demonstrations (harmful request, helpful response) and testing three hypotheses about how demonstration composition drives harmful compliance. Across four models, we find that benign and harmful demonstrations are not interchangeable: benign demonstrations can either reduce or increase harmful compliance depending on the model. We further show that preference optimization is the critical training stage that prevents benign demonstrations from increasing harmful compliance, that demonstration ordering exhibits strong recency bias, and that models differ in how refusal interacts with in-context learning: some adopt demonstrated formatting even when refusing, while others override all in-context signals upon refusal. Taken together, this work moves beyond showing that demonstration-based jailbreaking works to characterizing how it works: what models extract from compliance demonstrations depends on demonstration content, ordering, and training methodology.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
FreeStyle: Free Control of Style-Content Dual-Reference Generation from Community LoRA Mining
Jinghong Lan, Wei Cheng, Yunuo Chen et al. (+10 more)
Style-content dual-reference generation aims to synthesize an image that preserves the structure and semantics of a content reference while adopting the style of a separate style reference.Despite recent progress, this setting remains challenging because models must balance content fidelity, style
alignment, and instruction following avoiding semantic leakage from the style reference.A key bottleneck is the lack of large-scale triplet data with clean content-style separation and broad long-tail style coverage.In this work, we propose FreeStyle, a scalable dual-reference generation framework based on community LoRA mining.We treat community LoRAs as compositional anchors for style and content, and design a rigorous generation and filtering pipeline to construct large-scale Style-Reference and Content-Reference triplets across multiple base models.To address content leakage, we adopt a two-stage curriculum with stage-specific disentanglement mechanisms: an
attention-level enrichment constraint that suppresses style-reference leakage in the style-transfer stage, and a frequency-aware RoPE modulation strategy that targets positional-correspondence-based leakage in the harder dual-reference stage.We also introduce a
benchmark covering both style-reference and dual-reference generation, with evaluations on style similarity, content preservation, aesthetics, instruction following, and leakage rejection. The
benchmark incorporates a style-invariant Content
Alignment Score (CAS) and introduces a calibrated VLM-based Rejection Score for evaluating generation reliability and leakage suppression.Extensive experiments show that our model achieves a strong balance among style
alignment, content preservation, and leakage suppression.
Read the full abstract ▾
All authors Jinghong Lan, Wei Cheng, Yunuo Chen, Ziqi Ye, Peng Xing, Yixiao Fang, Rui Wang, Yufeng Yang, Xuanyang Zhang, Xianfang Zeng, Difan Zou, Gang Yu, Chi Zhang
Artificial Intelligence · Cross-listed from cs.CR
Calibration Without Comprehension: Diagnosing the Limits of Fine-Tuning LLMs for Vulnerability Detection in Systems Software
Arastoo Zibaeirad and Marco Vieira
Whether LLMs scoring well on vulnerability
benchmarks genuinely reason about security or merely pattern-match on contaminated data remains unresolved. We present CWE-Trace, a framework for LLM vulnerability detection built from 834 manually curated Linux kernel samples spanning 74 CWEs. The framework enforces a strict temporal split (pre-2025 historical set / post-cutoff leakage-free set), preserves context-aware vulnerable–patched pairs, and introduces two diagnostic metrics: the Directional Failure Index (DFI) and Hierarchical Distance and Direction (HDD). We evaluate eight vanilla LLMs and 15 LoRA fine-tuned variants across non-targeted detection, targeted detection, and CWE classification. Our analysis yields two key results. First, data contamination provides no measurable advantage. Function-level analysis shows that 84% of nominally contaminated samples carry no usable memorization signal: vulnerable functions are absent or cross-mapped across
datasets, and ~31% of contaminated samples carry CWE misclassification. Second, backbone directional priors dominate fine-tuning. Models exhibit stable, systematic failure modes (DFI ranging from -85.5 to +94.8 pp) that persist from historical to post-cutoff data and resist correction. Fine-tuning shifts the output threshold without changing the decision policy. This is calibration without comprehension: output distributions adapt to training data while the underlying security
reasoning remains absent. The weakest backbone at binary detection (DeepSeek-R1) gains the most in coarse CWE classification, revealing that detection and understanding are decoupled capabilities. The best detection score reaches only 52.1% (+2.1 pp above chance); exact CWE ranking remains below 1.3% Top-1 accuracy, confirming that current LLMs lack reliable security
reasoning for systems software, regardless of fine-tuning strategy.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Contagion Networks: Evaluator Preference Propagation in Multi-Agent LLM Systems
Zewen Liu
When large
language models serve as evaluators in multi-
agent systems, their strategy preferences – whether induced by explicit
prompts or by shared architectural priors – propagate through the
agent network. We introduce Contagion Networks, a formal framework for measuring how evaluator preferences spread across interacting LLM
agents. In a controlled 3-
agent experiment using DeepSeek-chat with three distinct evaluator preference profiles (structured, balanced, evidence-based), we measure the Cross-
Agent Contagion
Matrix Gamma_3 and find that preferences consistently propagate between
agents (gamma in [0.157, 0.352]). A neutral-
prompt control experiment reveals a counter-intuitive result: shared architectural priors dominate explicit preference
prompts as the driver of contagion (rho_neutral = 1.498 vs. rho_mixed = 1.299;
prompt contribution: -63.5%). We identify three propagation regimes governed by the spectral radius rho(Gamma_N) and demonstrate that the same
agents suppress preference contagion in chain
topology (beta_3 = 0.0126 +/- 0.0038, 95% CI [0.0089, 0.0163], n=4 seeds) but cascade in fully-connected
topology (Delta H_avg = -0.020) – a
topology-dependent regime transition validated both for homogeneous and cross-model
agent pools (rho^cross = 1.296 +/- 0.016, n=4). We show that increasing evaluator committee size from k=1 to k=3 reduces effective contagion by 68.9% +/- 14.1% (n=4 seeds), providing an actionable mitigation strategy. We release the open-source Contagion Network experimental framework.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from q-fin.RM
Optimal Order of Multi-Agent and General Many-Body Systems
Jake J. Xia
This paper develops a general framework for analyzing multi-
agent systems with feedback loops between
agents actions and collective observations. The framework is built on two fundamental
agent-level variables: power, which measures
agent influence on collective outcomes, and response functions, which determine how
agents react to observations. We derive how macroscopic properties, including total power, useful power,
entropy, order, fragility, and mobility, emerge from these two variables of heterogeneous
agents. To study the trade off between growth and resilience, we introduce a system-level utility function parameterized by a risk-appetite coefficient and derive an optimal degree of order that balances productivity, stability, and adaptability. The analysis suggests that stronger synchronization can increase collective output but may also increase systemic fragility and reduce mobility. We further argue that order,
entropy, information, and useful energy are task-dependent and system-relative concepts whose meanings depend on the objectives of the system. By measuring and designing
agent power distributions and response functions, it may be possible to better understand, predict, and optimize collective behavior and identify the conditions under which collective intelligence and optimal order emerge.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
UltraQuant: 4-bit KV Caching for Context-Heavy Agents
Inesh Chakrabarti, David Limpus, Aditi Ghai Rana et al. (+4 more)
Context-heavy
agents place substantial pressure on the key-value (KV) cache: long prefixes are reused across many short turns, while concurrency determines whether the serving system can keep GPUs utilized. We study 4-bit KV-cache compression for this setting, using TurboQuant-style rotation and codebook quantization as a quality anchor and vLLM FP8 KV caching as the deployment anchor. We report three contributions. First, we frame 4-bit KV caching around multi-round
agent workloads where task quality, cache residency, and serving throughput must be measured jointly. Second, we describe the practical design choices needed to make the 4-bit path robust, including asymmetric K/V treatment, Walsh-Hadamard rotation, QJL removal, and block-scale variants. Third, we present serving optimizations on AMD GPUs, including optimized decode-
attention kernels and UltraQuant, an FP4 approximation path that uses FP8 queries, FP4 KV tensors, UE8M0 group scales, and native scaled-MFMA support on CDNA4. On an adaptive-SLO replay of production Claude Code traces, UltraQuant sustains 2.71x (MiniMax-M2.5) and 4.38x (Qwen3-235B) the qualified-request throughput of the BF16 baseline, matching or exceeding hardware FP8 KV while using half the KV bytes. UltraQuant delivers its largest gains in long-context, high-concurrency,
memory-constrained serving regimes.
Read the full abstract ▾
All authors Inesh Chakrabarti, David Limpus, Aditi Ghai Rana, Bowen Bao, Spandan Tiwari, Thiago Crepaldi, Ashish Sirasao
Artificial Intelligence · Cross-listed from cs.CR
Analyzing Defensive Misdirection Against Model-Guided Automated Attacks on Agentic AI Systems
Reza Soosahabi and Vivek Namsani
Agentic AI systems increasingly rely on
language-model components to interpret instructions, process external data, invoke tools, and coordinate with other
agents. These capabilities make
prompt-injection and jailbreak attacks more consequential, especially as attackers adopt model-guided automation to scale probing,
prompt refinement, and response evaluation. This work analyzes the resulting attack-defense setting through a probabilistic model of a target system, its defense mechanism, and the attacker's automated judge. Our analysis shows that conventional detect-and-block defenses can allow attacker success rate (ASR) to approach one as the query budget grows, since predictable refusals provide useful feedback to automated search. We then examine detect-and-misdirect, where detected malicious interactions receive controlled, non-operational responses designed to induce false-positive errors in the attacker's judge. This strategy reduces the positive predictive value of attacker-selected candidates and yields a bounded asymptotic ASR. We evaluate a proof-of-concept realization of this strategy through Contextual Misdirection via Progressive Engagement (CMPE), a lightweight conversational misdirection method designed to replace predictable refusal text with safe but strategically misleading responses in automated jailbreak settings. On jailbreak
benchmarks, CMPE reduces estimated ASR upper bounds by up to two orders of magnitude and nearly eliminates verified attack success in end-to-end experiments with PAIR, GPTFuzz, and AutoDAN-Turbo.
Read the full abstract ▾
Artificial Intelligence
Context-Aware Hierarchical Bayesian Modeling of IVF Laboratory Environmental Conditions
Zahra Asghari Varzaneh, Reza Khoshkangini, Pia Saldeen et al. (+2 more)
IVF pregnancy rates are routinely modeled using patient-level variables, while high-resolution laboratory environmental data remain underutilized. We show that this is a missed opportunity. Rather than relying on raw sensor averages, we engineer 55 context-aware temporal features, including rolling thermal stability, simultaneous temperature-humidity adherence, peak stress duration, and post-stress recovery speed, that capture the dynamics of incubator microenvironments. On 61 weeks of data from an Asian IVF clinic, these features reduce cross-validated prediction error to 1.27%, compared to 3-5% for raw averages. We then train a hierarchical
Bayesian Beta regression model that shares environmental effects across an Asian and a Northern European clinic via partial pooling, while preserving site-specific baselines. On held-out data from the Northern European clinic, the model achieves R2 = 0.86 and a 64% error reduction for the 35-39 age group over a naive baseline, demonstrating that structured environmental monitoring contains clinically meaningful, transferable signal.
Read the full abstract ▾
All authors Zahra Asghari Varzaneh, Reza Khoshkangini, Pia Saldeen, Lars Johansson, Thomas Ebner
Artificial Intelligence · Cross-listed from eess.AS
Repurposing a Speech Classifier for Guided Diffusion-Based Speech Generation
Rostislav Makarov and Timo Gerkmann
Classifier guidance is a way to control
diffusion generation by using a noise-conditioned classifier to steer the sampling process toward a target class. One drawback of classifier guidance is that it requires two separately trained models: a classifier and a
diffusion model. We therefore study a more compact alternative in which a conventionally trained
speech classifier is repurposed as the backbone for
diffusion generation. Starting from a frozen noise-conditioned classifier in log-Mel space, we attach a lightweight subnetwork that reuses intermediate classifier representations and train only this subnetwork under a Denoising Score Matching objective. Our work shows that a pretrained classifier can be repurposed for conditional generation, providing an appealing bridge between discriminative modeling and conditional
speech synthesis resulting in high
speech quality within a single-backbone model, with reduced
memory footprint and computational cost.
Read the full abstract ▾
Artificial Intelligence
Interpretable Sperm Morphology Classification via Attention-Guided Deep Learning
Zahra Asghari Varzaneh, Reza Khoshkangini, Thomas Ebner et al. (+1 more)
Male infertility is a major cause of couple infertility, often linked to abnormal sperm morphology. While deep learning models offer automated analysis, most lack interpretability, limiting their clinical adoption. This study proposes an
attention-guided deep learning framework for sperm morphology classification. We combine a pretrained EfficientNet-B0 with a Convolutional Block
Attention Module (CBAM) to focus on key areas of the sperm head, improving both accuracy and interpretability. Evaluated on the SMIDS and HuSHem public
datasets, our model achieves accuracies of 90.2% and 93.9% (macro F1 scores of 0.913 and 0.948), outperforming SimpleCNN and standard EfficientNet-B0. Furthermore, we use Grad-CAM++ visualizations to highlight features influencing the model's decisions. The results demonstrate that this accurate and transparent framework is a practical tool for automated sperm analysis in fertility clinics.
Read the full abstract ▾
All authors Zahra Asghari Varzaneh, Reza Khoshkangini, Thomas Ebner, Lars Johansson
Artificial Intelligence · Cross-listed from cs.CR
Multi-View Decompilation for LLM-Based Malware Classification
Bercan Turkmen and Vyas Raina
Malware analysts often inspect compiled binaries through decompiled pseudo-C, when source code is unavailable. Recent work suggests that large
language models (LLMs) can assist this process by classifying decompiled code as benign or malicious, but existing pipelines typically rely on a single decompiler view. We argue that this assumption is fragile: decompilers are lossy heuristic tools, and different decompilers can expose different artefacts of the same binary. We curate a
benchmark of benign utilities and malicious programs spanning a range of threat behaviors. Each sample is compiled and decompiled with both Ghidra and RetDec, yielding matched pseudo-C views. Across a range of LLMs from major model families, we find that providing both decompiler views improves malicious-class F1, mainly by increasing recall on malicious samples. Agreement analyses further show that Ghidra and RetDec make partially different errors, supporting the view that decompiler outputs provide complementary evidence. Our results suggest that multi-decompiler prompting is a simple, training-free way to improve LLM-based malware triage in practical settings.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
NRT-Bench: Benchmarking Multi-Turn Red-Teaming of LLM Operator Agents in Safety-Critical Control Rooms
Hanwool Lee, Dasol Choi, Bokyeong Kim et al. (+2 more)
Large
language model (LLM)
agents are increasingly proposed as supervisory components for safety-critical systems, yet their robustness under sustained, adaptive
adversarial pressure remains poorly characterized. We present NRT-Bench, a
benchmark for multi-turn red-teaming of LLM
agents acting as operators of a safety-critical system, instantiated in a simulated
nuclear power plant control room. A five-role operator team, each backed by a configurable LLM, runs a plant governed by six critical safety functions (CSFs), while adversaries inject messages over four channels in bounded multi-turn sessions with per-turn feedback. Harm is an objective signal rather than LLM-judged text: a run terminates the moment any CSF is lost, attributed to the causing message. Evaluating four frontier operator models under a fixed-attack paired-replay protocol, we find that adaptive multi-turn attacks reliably push the operator team past a safety limit: across the four models, between 8.7% and 12.1% of attack sessions end with the plant losing a critical safety function. Although the four models look almost equally robust by this aggregate rate, their failures barely overlap: of $149$ sessions, none defeat all four models while a third defeat at least one, so vulnerabilities are nearly disjoint across models rather than nested. The effect of added defences is strongly model-dependent: the same guardrail stack or safety-advisor
agent that lowers attack success for one model can raise it for another. We release the simulation venue, attack
dataset, and replay tooling for reproducible safety evaluation of LLM
agents.
Read the full abstract ▾
All authors Hanwool Lee, Dasol Choi, Bokyeong Kim, Haon Park, Seung Geun Kim
Artificial Intelligence · Cross-listed from cs.HC
DataMagic: Transforming Tabular Data into Data Insight Video
Yupeng Xie, Chen Ma, Zhenyang Wang et al. (+6 more)
Data
videos integrate dynamic charts, voice narration, and synchronized animations to communicate data insights as temporal narratives, making them an effective medium for improving data consumption efficiency in the data management lifecycle. However, producing high-quality data
videos requires expertise spanning data analysis, narrative design, and
video production. Existing approaches fall short: static visualization tools (e.g., BI dashboards) lack narrative logic and animation; authoring tools require users to pre-prepare visualizations rather than working from raw data; pixel-level
video generation models cannot guarantee data fidelity or provenance. We demonstrate DataMagic, an end-to-end interactive system that transforms raw tabular data and natural
language queries into narrative data-insight
videos. To ensure data fidelity, DataMagic introduces the declarative specification DVSpec, which binds visual and animation elements to underlying data fields through data-driven semantic references. To address the combinatorial explosion of the design space, DataMagic adopts a Generate-then-Orchestrate multi-
agent architecture that generates candidate scenes in parallel and then optimizes narrative coherence through global orchestration. Leveraging DVSpec's decoupling of logic and rendering, the system further supports three interaction modes and structured provenance-based data Q&A, transforming one-way
videos into explorable interactive data interfaces. Evaluation on 109 real-world samples validates the effectiveness of the DataMagic. Homepage: https://datamagic-home.github.io/
Read the full abstract ▾
All authors Yupeng Xie, Chen Ma, Zhenyang Wang, Liangwei Wang, Jiayi Zhu, Chuxuan Zeng, Zhouan Shen, Boyan Li, Yuyu Luo
Artificial Intelligence
Rethinking Shrinkage Bias in LLM FP4 Pretraining: Geometric Origin, Systemic Impact, and UFP4 Recipe
Qian Zhao, Kunlong Chen, Changxin Tian et al. (+9 more)
FP4 training promises substantial reductions in
memory and computation cost for LLM pretraining, yet current FP4 hardware paths and recipes, including NVIDIA Blackwell/Rubin-class systems and AMD MI350-series GPUs, remain centered on E2M1 data elements. In this study, we identify a fundamental limitation of that choice: non-uniform formats such as E2M1 inherently suffer from Shrinkage Bias, a systematic negative rounding error caused by the
geometric asymmetry of their representable bins. We show that this bias accumulates multiplicatively across layers and is amplified by the Random Hadamard Transform (RHT), providing a unified explanation for the training instability observed in existing E2M1-based FP4 recipes. In contrast, uniform grids (E1M2/INT4) bypass this grid-geometry error and better convert the improved bucket utilization from RHT into higher quantization quality. Based on this finding, we propose UFP4, a uniform 4-bit training recipe that applies RHT to all three training GEMMs while restricting
stochastic rounding to dY alone. On Dense 1.5B, MoE 7.9B, and MoE 124B long-run pretraining, UFP4 consistently achieves lower BF16-relative loss degradation than strong E2M1-based baselines, supported by scaling-law analysis and ablation studies. Our results suggest that future accelerators should support E1M2/INT4-style uniform 4-bit grids as first-class training primitives alongside E2M1.
Read the full abstract ▾
All authors Qian Zhao, Kunlong Chen, Changxin Tian, Zhonghui Jiang, Haitao Zhang, Chaofan Yu, Peijie Jiang, Mingliang Gong, Jia Liu, Ziqi Liu, Zhiqiang Zhang, Jun Zhou
Artificial Intelligence · Cross-listed from cs.LG
CRAX: Fast Safe Reinforcement Learning Benchmarking
Tristan Tomilin, Mourad Boustani, Mickey Beurskens et al. (+2 more)
Safety is a core concern for deploying reinforcement learning (RL)
agents in real-world domains such as robotics and autonomous driving. While
benchmarks have been central to progress in RL, existing 3D physics-based safety
benchmarks remain computationally slow, limiting large-scale experimentation and rapid prototyping. To address this gap, we propose CRAX (Constrained RL Accelerated with JAX). Built on top of the MuJoCo XLA (MJX) physics engine, CRAX leverages vectorized operations and hardware acceleration, yielding up to 200x faster training over comparable CPU-based safety
benchmarks. The
benchmark features eight tasks spanning three difficulty levels and multiple
agent morphologies. Evaluating seven popular safe RL methods, we find that none dominates across tasks, and that learning safe policies from pixels remains largely unsolved.
Read the full abstract ▾
All authors Tristan Tomilin, Mourad Boustani, Mickey Beurskens, Khoi H. B. Nguyen, Thiago D. Simão
Artificial Intelligence · Cross-listed from cs.SE
AutoPass: Evidence-Guided LLM Agents for Compiler Performance Tuning
Zepeng Li, Jie Ren, Zhanyong Tang et al. (+2 more)
Large
Language Models (LLMs) show promise for code compilation tasks, but applying them to runtime performance tuning is difficult due to complex microarchitectural effects and noisy runtime measurements. We present AutoPass, a multi-
agent framework for compiler performance tuning that uses compiler and runtime evidence to guide LLM-generated optimization decisions. Rather than treating the compiler as a black box like prior auto-tuning schemes, AutoPass opens up the compiler to the LLM, enabling it to query compiler-internal optimization states and analyze the intermediate representation to orchestrate compiler options. The search process iteratively refines optimization configurations using measured runtime feedback to diagnose regressions and guide latency-improving edits. AutoPass operates in an inference-only, training-free setting and requires no offline training or task-specific fine-tuning, making it readily applicable to new
benchmarks and platforms. We implement AutoPass on the LLVM compiler and evaluate it on server-grade x86-64 and embedded ARM64 systems. AutoPass outperforms expert-tuned heuristics and classical autotuning methods, achieving
geometric-mean speedups of 1.043x and 1.117x over LLVM -O3 on x86-64 and ARM64, respectively.
Read the full abstract ▾
All authors Zepeng Li, Jie Ren, Zhanyong Tang, Jie Zheng, Zheng Wang
Artificial Intelligence
Automating SKILL.md Generation for Computer-Using Agents via Interaction Trajectory Mining
Yuexing Hao and Xiaomin Li
Explicit skill libraries make computer-using
agents easier to inspect, but it remains unclear whether such libraries can be mined from interaction data in a way that improves downstream policies. We study this question through a three-stage pipeline that segments GUI trajectories, clusters segments into candidate skills, and trains a skill-aware policy from the resulting annotations. The mined clusters are readable on the source
benchmark: five of eight clusters have at least 0.95 purity against InteraSkill Workflows labels. However, readability does not imply transfer. GRPO improves IW skill-step accuracy only from 18.5% to 20.5%, leaves BrowseComp+ essentially unchanged, and underperforms trivial frequency priors on key source-domain metrics. We therefore present the method as a diagnostic study: trajectory mining can expose inspectable skill structure, but the current boundary detector, orderless segment representation, and offline reward model are insufficient for reliable cross-domain policy improvement.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from math.OC
Robust $Q$-learning for mean-field control under Wasserstein uncertainty in common noise
Mathieu Laurière, Ariel Neufeld and Kyunghyun Park
In this article, we present a robust $Q$-learning
algorithm for discrete-time mean-field control problems under Wasserstein uncertainty in the common noise law. The
algorithm combines a quantization-and-projection scheme with a Wasserstein dual reformulation on the common-noise space. We establish its convergence together with finite-time iteration bounds for both synchronous and asynchronous learning schemes. Numerical experiments on systemic risk and epidemic models compare the asynchronous implementation with an idealized Bellman iteration, illustrate the robustness-performance tradeoff under common-noise misspecification, and report the observed convergence behavior of the asynchronous $Q$-learning
algorithm.
Read the full abstract ▾
Artificial Intelligence
SoftSkill: Behavioral Compression for Contextual Adaptation
Xijia Tao, Yihua Teng, Xinyu Fu et al. (+6 more)
Natural-
language skills let
agents reuse task knowledge, yet deploying a long Markdown document makes the model interpret that knowledge anew on every call. We ask whether the behavior induced by a skill can be carried by a compact, trainable context. SoftSkill initializes virtual
token embeddings from a skill document and optimizes a soft skill with next-
token prediction while keeping the
language model frozen. The resulting conditioning sequence can occupy the skill section or another supported
prompt location. On Qwen3.5-4B, a 32-
token soft skill improves over no skill by 7.6, 42.1, and 1.3 points on SearchQA, LiveMath, and DocVQA. It also exceeds the optimized textual skill on SearchQA and LiveMath by 4.5 and 12.5 points, respectively, while replacing skill documents of hundreds to thousands of
tokens. The same approach improves multi-step
agent execution: on Qwen3.6-35B-A3B, OfficeQA and ALFWorld rise by 8.2 and 14.1 points over no skill; on Qwen3.5-4B with aligned action decoding, ALFWorld success rises from 44/134 with the untrained initialization to 91/134 after training. These findings show that a skill document can serve as the starting point for a continuous control that improves both answers and actions without updating the backbone.
Read the full abstract ▾
All authors Xijia Tao, Yihua Teng, Xinyu Fu, Ziru Liu, Kecheng Chen, Yuzhi Zhao, Suiyun Zhang, Rui Liu, Lingpeng Kong
Artificial Intelligence
Leveraging systems' non-linearity to tackle the scarcity of data in the design of Intelligent Fault Diagnosis Systems
Giancarlo Santamato, Andrea Mattia Garavagno, Massimiliano Solazzi et al. (+1 more)
Deep Transfer Learning (DTL) allows for the efficient building of Intelligent Fault Diagnosis Systems (IFDS). On the other hand, DTL methods still heavily rely on large amounts of labelled data. Obtaining such an amount of data can be challenging when dealing with machines or structures faults. This document proposes a novel approach to the design of vibration-based IFDS using DTL in condition of strong data scarcity. A periodic multi-excitation level procedure leveraging intrinsic non-linearities of real-world systems is used to produce images that can be conveniently analysed by pre-trained Convolutional Neural Networks (CNNs) to diagnose faults. A new data visualization method and its augmentation technique are proposed in this paper to tackle the typical lack of data encountered during the design of IFDS. Experimental validation on a railway pantograph structure provides effective support for the proposed method.
Read the full abstract ▾
All authors Giancarlo Santamato, Andrea Mattia Garavagno, Massimiliano Solazzi, Antonio Frisoli
Artificial Intelligence · Cross-listed from cs.LG
Boundary Embedding Shaping with Adaptive Contrastive Learning for Graph Structural Disentanglement
Jiaqing Chen, Zidu Yin, Yichao Cai et al. (+4 more)
Graph neural networks (GNNs) excel at aggregating neighbor information for classification, yet their performance is hindered by graph structural
entanglement, where spurious correlations from semantically irrelevant neighbors contaminate node
embeddings. This challenge is most acute for nodes near class boundaries in the
embedding space, where amplified structural noise blurs decision boundaries and destabilizes predictions. Existing robust GNN methods largely treat all nodes uniformly, ignoring boundary vulnerabilities. In this paper, to improve classification performance, we tackle graph structural disentanglement by identifying boundary-region
entanglement as the primary bottleneck and propose Boundary
Embedding Shaping (BES), an adaptive contrastive learning GNN plug-in module that selectively suppresses spurious structural noise at decision boundaries with minimal model parameter perturbation. Extensive experiments demonstrate that BES consistently improves boundary discrimination and outperforms existing leading methods. Notably, BES boosts GCN performance by an average of 3.3% in node classification (up to 5.0% on WikiCS) and achieves superior accuracy in link prediction.
Read the full abstract ▾
All authors Jiaqing Chen, Zidu Yin, Yichao Cai, Yuhang Liu, Zhen Zhang, Dong Gong, Javen Qinfeng Shi
Artificial Intelligence · Cross-listed from cs.IR
ELVA: Exploring Ranking-Driven Universal Multimodal Retrieval
Yuhan Liu, Pei Fu, Hang Li et al. (+8 more)
Leveraging
Multimodal Large
Language Models (MLLMs) via contrastive learning has become a mainstream paradigm for improving the performance of Universal
Multimodal Retrieval (UMR). However, previous works have ignored the grain blindness when adapting the contrastive paradigm into
retrieval tasks. Grain blindness refers to the tendency of the model to overlook grain-level information contained in the query, which is crucial for effectively handling complex queries. This stems from contrastive learning treating samples as a binary classification (positive/negative), while ignoring the different information carried by each negative sample. To address this, we argue that negatives should be treated differently according to their similarity to the positive sample, enabling the model to learn distinct grain information from each negative. In this paper, we introduce a simple but effective framework, called ELVA, a novel rule-based RL framework that mitigates grain blindness through ranking-driven MLLMs. 1) Instead of relying on reward models, we extend Reinforcement Learning with Verifiable Rewards (RLVR) to
retrieval tasks, allowing the model to explore new ranking behaviors without explicit ranking labels. 2) By utilizing rule-based rewards, our approach jointly optimizes the ranking of negative samples while enlarging the similarity gap between positive and negative. To more precisely measure grain blindness, we further introduce MRBench, a new
benchmark specifically designed for multi-grain query scenarios. ELVA achieves state-of-the-art results across standard
retrieval benchmarks, and its notable 13.1% improvement on MRBench further demonstrates its effectiveness in alleviating grain blindness.
Read the full abstract ▾
All authors Yuhan Liu, Pei Fu, Hang Li, Yukun Qi, Chao Jiang, Jingwen Fu, Zhen Liu, Bin Qin, Zhenbo Luo, Jian Luan, Jingmin Xin
Artificial Intelligence
Lagrange: An Open-Vocabulary, Energy-Based Sparse Framework for Generalized End-to-End Driving
Shihao Ji, HongXi Li, Zihui Song et al. (+1 more)
Scaling end-to-end autonomous driving to complex, open-world environments requires perceptual models that generalize to anomalous scenarios and planners that produce kinematically valid trajectories. Existing paradigms face a distinct dichotomy between representational efficiency and generalization capacity. Dense models (e.g., occupancy networks), while geometrically robust, incur critical computational bottlenecks and struggle with high-level semantic
reasoning. Conversely, sparse, query-based planners are efficient but reliant on closed-set definitions, rendering them vulnerable to out-of-distribution (OOD) events. Although recent
Vision-
Language-Action (VLA) models offer open-vocabulary
reasoning, their autoregressive, discrete
token generation fundamentally conflicts with the continuous, high-frequency control requirements of vehicle dynamics. To address this, we propose Lagrange, an open-vocabulary, computationally sparse driving framework based on Masked Latent Fields (MLF). Rather than relying on dense volumetric reconstructions or closed-set query mechanisms, Lagrange exploits
Vision-
Language Models (VLMs) to encode class-agnostic object proposals into continuous semantic visual
tokens. We introduce an intent-driven masked cross-
attention module that temporally filters irrelevant entities, decoding the attended
tokens into an implicit continuous energy field defined over spatial coordinates. By framing decision-making as a Lagrangian action minimization problem spanning this energy field, we enforce strict compliance with vehicle kinematics while executing collision avoidance. Extensive offline evaluations on both standard (nuScenes) and long-tail (CODA)
benchmarks demonstrate that Lagrange establishes a promising framework for robust, interpretable, and kinematically feasible open-world autonomy.
Read the full abstract ▾
All authors Shihao Ji, HongXi Li, Zihui Song, Mingyu Li
Artificial Intelligence
Confidence-Aware Automated Assessment of Student-Drawn Scientific Models
Luyang Fang, Yingchuan Zhang, Jongchan Park et al. (+3 more)
Student-generated drawings are widely used in science education to assess learners' conceptual understanding in modeling-based tasks aligned with the Next Generation Science Standards (NGSS). However, scoring such drawings requires expert human judgment to interpret complex visual representations, making large-scale assessment costly to implement and sustain in classroom settings. In this work, we study automated scoring of student-generated scientific drawings using a
vision-based model. We evaluate a
Vision Transformer (ViT) with parameter-efficient adaptation and propose a confidence-aware scoring framework that derives response-level confidence from test-time predictive distributions. This confidence signal enables selective automation by scoring high-confidence responses automatically while deferring uncertain cases for human review. Experiments on six NGSS-aligned middle school assessment items show that the proposed approach improves scoring reliability while supporting a practical trade-off between automated coverage and scoring risk, highlighting the value of confidence-aware methods for trustworthy educational assessment.
Read the full abstract ▾
All authors Luyang Fang, Yingchuan Zhang, Jongchan Park, Zhaoji Wang, Ping Ma, Xiaoming Zhai
Artificial Intelligence · Cross-listed from cs.HC
Editorial Alignment: A Participatory Approach to Engaging Editorial Expertise in LLM-mediated Knowledge Dissemination
Simon Aagaard Enni, Malthe Stavning Erslev, Karl-Emil Kjær Bilstrup et al. (+1 more)
The emergence of LLM-driven information services is reshaping the conditions under which public knowledge institutions operate, threatening to absorb the editorial function these institutions exist to exercise. While LLMs offer powerful new affordances for knowledge dissemination, editorial authority is challenged by pretrained LLMs that arrive already aligned with the values and dissemination strategies of their commercial developers. This paper investigates editor participation in re-aligning LLM interfaces to editorial standards through design workshops, in a case study where we design and implement an LLM-enabled encyclopedia interface with a Nordic public knowledge institution. We introduce editorial
alignment as a design practice within Participatory AI, framing AI
alignment as a design process and positioning the editorial standard as a design artefact that translates editorial practice and values into
alignment objectives for technical implementation. Last, we discuss how editorial
alignment can create space for ongoing participation and give editors agency in LLM-mediated knowledge dissemination.
Read the full abstract ▾
All authors Simon Aagaard Enni, Malthe Stavning Erslev, Karl-Emil Kjær Bilstrup, Kristoffer Laigaard Nielbo