Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

01.
arXiv (CS.AI) 2026-06-19

AURA: Adaptive Uncertainty-aware Refinement for LLM-as-a-Judge Auditing

arXiv:2606.19714v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as judges for open-ended generation, as large-scale human evaluation is often expensive and difficult to scale, yet their preferences remain imperfect proxies for human judgment. Existing auditing pipelines often assume that a reliable subset of examples or clean supervision signals are available beforehand, for example from human annotation, heuristic filtering, or the outputs of strong judges. In LLM evaluation, this assumption is fragile: the initial split may inherit judge bias, while human verification is typically too scarce to define stable groups at scale. We propose AURA, an adaptive uncertainty–aware refinement framework for auditing pairwise LLM–as–a–judge decisions under selected human verification. AURA iteratively learns a human-consistency signal, propagates reliable evidence, and prioritizes uncertain comparisons for human review. The key idea is to treat trust in a judge as a latent quantity that is progressively refined as evidence accumulates. We provide a compact formulation, a stable refinement procedure, and a comprehensive evaluation on both synthetic and real pairwise LLM-answer data.

02.
arXiv (CS.AI) 2026-06-18

Learning-Based Decision Making for Combustion Phasing Control in Multi-Fuel CI Engines with Latent Fuel Reactivity Estimation

arXiv:2606.18393v1 Announce Type: cross Abstract: Multi-fuel compression-ignition engines offer fuel flexibility but introduce uncertain, time-varying fuel reactivity, represented by cetane number (CN), which complicates cycle-to-cycle combustion-phasing control. This work formulates CA50 regulation under latent CN variation as a partially observable sequential decision problem and systematically evaluates controllers with increasing temporal and representational capacity, including LinUCB, history-augmented contextual bandits, observation-only DDPG, recurrent DDPG, and a proposed GRU-guided RL framework. A Gaussian-process surrogate trained on experimental multi-fuel engine data provides a controlled and reproducible evaluation environment. Results show that myopic and fixed-history bandit methods degrade under CN variation, observation-only RL suffers from latent-state aliasing, and generic recurrence is insufficient when CN evolves rapidly. The proposed framework learns a compact GRU-based representation of fuel reactivity from combustion history and conditions both actor and critic on this estimated signal rather than oracle CN. By training the policy on the same imperfect fuel-reactivity information available at deployment, the controller avoids train-deploy inconsistency in conventional online estimate-then-control pipelines. Across unseen CN trajectories, the policy achieves stable CA50 regulation with mean absolute tracking error below 0.25{\deg} CA at the training setpoint, while producing smooth, physically consistent SOI and glow-plug-power actuation. These results show that combustion control under latent, continuously evolving fuel dynamics requires more than standalone estimation or generic recurrence. By aligning fuel-reactivity inference with control policy learning, the proposed framework enables reactivity-aware decision-making using the same estimated state available during deployment.

03.
arXiv (CS.CV) 2026-06-12

VISTA: An End-to-End Benchmark for Visual Spec-to-Web-App Coding Agents

We present VISTA (VIsual Spec-To-App Benchmark), a benchmark for evaluating the end-to-end web-app generation capabilities of LLM-based agents. Unlike prior code generation benchmarks that focus on algorithmic tasks, VISTA targets realistic UI-centric development, where agents must produce functional, visually coherent applications from underspecified inputs. We define five prompt-information conditions that vary along two axes, visual/structural fidelity and stack constraint: (1) text only with free stack choice, (2) text with reference screenshots under three specified stacks, (3) text with reference screenshots under free stack choice, (4) text with screenshots and pruned Figma structure under a single specified stack, and (5) text with screenshots and pruned Figma structure under free stack choice. To enable robust evaluation, each page in the benchmark is manually annotated with interactive UI components and around three visual anchor points, addressing the well-known limitations of script-based testing tools such as Playwright in open-ended code generation settings. Evaluation combines DOM-grounded reference matching, behavior-specific browser tests, and CLIP-based visual similarity, jointly measuring structural alignment, behavioral completeness, and overall visual fidelity. We use VISTA to assess four agent systems drawn from two model families and two harnesses, finding that visual fidelity and functional correctness are partially decoupled across both input conditions and agents, and that agent editing style varies sharply but is largely orthogonal to task quality. VISTA establishes a rigorous and reproducible foundation for advancing agent-based software engineering research.

04.
arXiv (quant-ph) 2026-06-12

Coupling-Grouped XY-QAOA for Joint Anomaly-Feature Selection

arXiv:2606.13244v1 Announce Type: new Abstract: Selecting anomalous samples and explanatory features under fixed budgets defines a coupled constrained-optimization problem. Sequential feature-first selection ranks features before choosing samples, which can overlook features whose utility depends on which samples are selected, especially when scores are calibrated from reference data that may be limited, noisy, or drifting. We instead formulate the task as joint sample-feature selection under the same fixed counts. In the analyzed formal model, calibration-error sensitivity grows linearly with the number of samples for feature-first ordering but stays constant for joint selection. We introduce Coupling-Grouped XY-QAOA, a constraint-preserving grouped-angle variant for the resulting optimization problem. On matched sparse IBM Heron R3 benchmarks, a hardware-aware implementation reduces circuit depth by 45.9%-61.3% and two-qubit gates by 2.6%-5.2% relative to Qiskit optimization level 3 on the CZ-basis target. It enables, to our knowledge, the largest reported width-depth configurations for constraint-preserving bipartite-selection QAOA hardware executions with feasible-sector retention: 64 qubits at p=2 and 36 qubits at p=3. The 20-qubit p=5 runs retain 63% valid samples. Across 36-64 qubits, fixed-angle runs yield lower-energy feasible samples than matched random-feasible sampling. Warm starts reduce the gap to strict-feasible classical references by 57.5%-80.5%, and near-budget repair matches the sparse classical reference at 36 qubits. Benchmarks show gains in balanced fixed-budget regimes, and noiseless simulations show that problem-structured angle grouping improves over same-depth XY-QAOA and matched-parameter, type-preserving randomization controls. Overall, the results support calibrated joint selection and hardware-realizable constrained-mixer execution in the tested regimes.

05.
arXiv (CS.LG) 2026-06-16

ML Inference Scheduling with Predictable Latency

arXiv:2512.18725v3 Announce Type: replace Abstract: Machine learning (ML) inference serving systems can schedule requests to improve GPU utilization and to meet service level objectives (SLOs) or deadlines. However, improving GPU utilization may compromise latency-sensitive scheduling, as concurrent tasks contend for GPU resources and thereby introduce interference. Given that interference effects introduce unpredictability in scheduling, neglecting them may compromise SLO or deadline satisfaction. Nevertheless, existing interference prediction approaches remain limited in several respects, which may restrict their usefulness for scheduling. First, they are often coarse-grained, which ignores runtime co-location dynamics and thus restricts their accuracy in interference prediction. Second, they tend to use a static prediction model, which may not effectively cope with different workload characteristics. In this paper, we evaluate the potential limitations of existing interference prediction approaches, finding that coarse-grained methods can lead to noticeable deviations in prediction accuracy and that static models degrade considerably under changing workloads.

06.
bioRxiv (Bioinfo) 2026-06-16

PhenoBIC: operator-free single-cell spatial phenotyping in multiplex imaging data using deep learning of cell staining patterns

Multiplex imaging is a valuable tool for spatially examining tissue microenvironments at the single-cell level to uncover biological and clinical insights. However, most multiplex image analysis workflows currently require manual intervention for cell phenotyping, which slows progress, demands human effort, and yields operator-dependent outputs. Here, we developed PhenoBIC, a pre-trained deep learning model for image classification of the multiplexed biomarker signals in a cell (Biomarker Imprint of a Cell) to classify cell phenotypes. We show that PhenoBIC (F1-score ~0.88) outperforms manual gating (widely used) and other machine learning-based computational approaches for cell marker expression classification. We validated this across multiple biomarkers, tissue sampling strategies (whole biopsies and tissue microarrays), multiplex panels, imaging platforms, and tissue types. We have released our in-house training and validation datasets of ~1.4 million manually curated cell expression ground truth labels. We have also open-sourced PhenoBIC and enabled its community-wide deployment via the QuPath interface.

07.
arXiv (CS.AI) 2026-06-11

When Researchers Say Mental Model/Theory of Mind of AI, What Are They Really Talking About?

arXiv:2510.02660v2 Announce Type: replace-cross Abstract: When researchers claim AI systems possess ToM or mental models, they are fundamentally discussing behavioral predictions and bias corrections rather than genuine mental states. This position paper argues that the current discourse conflates sophisticated pattern matching with authentic cognition, missing a crucial distinction between simulation and experience. While recent studies show LLMs achieving human-level performance on ToM laboratory tasks, these results are based only on behavioral mimicry. More importantly, the entire testing paradigm may be flawed in applying individual human cognitive tests to AI systems, but assessing human cognition directly in the moment of human-AI interaction. I suggest shifting focus toward mutual ToM frameworks that acknowledge the simultaneous contributions of human cognition and AI algorithms, emphasizing the interaction dynamics, instead of testing AI in isolation.

08.
arXiv (CS.LG) 2026-06-15

Traditional machine learning vs. deep learning from dynamic graph representations of proteins' 3D folds in the task of protein structure classification

arXiv:2605.29228v2 Announce Type: replace Abstract: Protein structure classification (PSC) uses supervised learning to predict a protein's CATH/SCOP(e) class from the protein's sequence or 3D structural feature(s). We already modeled 3D structures as (static) protein structure networks (PSNs), demonstrating the competitiveness of PSN-based features to sequence or direct (i.e. non-network) 3D structural features in the PSC task. More recently, we demonstrated the power of features extracted from dynamic PSNs over features extracted from static PSNs (and thus by transitivity over sequence and direct 3D structural features) in the same task. That dynamic PSN approach used traditional machine learning (ML), combining manual (pre-engineered) features with an off-the-shelf classifier. Here, we evaluate whether automatic deep learning (DL) from the dynamic PSNs yields improvements. Our evaluation on 72 datasets spanning ~44,000 CATH- or SCOPe-labeled dynamic PSNs reveals that in terms of PSC accuracy, traditional ML and DL are (close to) tied for a large majority of the datasets, while DL is on average 10+ times slower. We are the first to evaluate traditional ML vs. DL in the dynamic PSN-based PSC task.

10.
PLOS Medicine 2026-05-22

Differences in tuberculosis prevalence by sex in low- and middle-income countries over 1993–2025: A systematic review and meta-analysis

by Nicole A. Swartwood, Nanki Singh, Seyed Alireza Mortazavi, Melike Hazal Can, Hening Cui, Do Kyung Ryuk, Peter MacPherson, Katherine C. Horton, Nicolas A. Menzies Background Global and national initiatives to combat tuberculosis (TB) have expanded over recent years. Despite this, the TB burden remains high in some population groups, with men recognized as having elevated TB risks. Summary measures of sex differences in TB prevalence were last estimated in 2016. Since then, many additional prevalence surveys have been conducted, including in the highest TB burden countries. We conducted a systematic review of sex-stratified TB prevalence survey data published over 1993–2025, to provide updated estimates of male-to-female (M:F) TB prevalence ratios and determine whether sex-related disparities in TB burden have closed over time. Methods and findings We identified surveys reporting community-representative, sex-stratified estimates of pulmonary TB prevalence in low- and middle-income countries (LMICs), including surveys from an earlier review (covering January 1993–March 2016) and a new systematic review (covering 1st December 2015–13th October 2025). This review was prospectively registered with PROSPERO (CRD42024503853) and included searches of PubMed, Embase, Global Health, the Cochrane Library, Africa Index Medicus, LILACS, and SciELO. We extracted data on bacteriologically confirmed and smear-positive TB prevalence among adults (aged ≥ 15 years), stratified by sex. Risk of bias was evaluated using eight criteria specific to prevalence surveys. We fit multi-level Bayesian regression models with study- and country-level random effects to estimate the M:F ratio of TB prevalence (male prevalence divided by female prevalence), overall and for key subgroups. In meta-regression analyses, we estimated how prevalence ratios varied over time and according to known TB risk factors and TB case definitions.We identified 10,124 publications and extracted data from 100 eligible studies representing 102 unique prevalence surveys and 4,658,310 participants (45.6% male) in 33 LMICs. TB prevalence was higher in men than women in 90/102 of the included surveys, with a pooled M:F prevalence ratio of 2.02 (95% credible interval (CrI): 1.71, 2.34) for bacteriologically confirmed TB and 2.38 (95% CrI: 1.91, 2.90) for smear-positive TB. Time trend analyses showed a 2.0% (95% CrI: −0.2, 4.5%) average annual change in the M:F ratio of bacteriologically confirmed TB over the study period. The M:F prevalence ratio was estimated to be higher for countries with greater excess HIV prevalence among men, and countries with greater gender equity (as measured by the United Nation’s Gender Development Index). The estimated M:F prevalence ratio was also higher for surveys that did not restrict testing to individuals reporting TB symptoms. Study limitations include heterogeneity in survey methods and definitions, as well as limited data from the Americas, Eastern Mediterranean, and Europe WHO world regions and post-COVID-19 period. Conclusions Men in LMICs consistently experience TB at a higher prevalence than women. Time trend estimates are uncertain, but consistent with widening sex differences in TB prevalence over the last three decades, despite efforts to address the risk factors underlying this excess TB burden.

11.
arXiv (quant-ph) 2026-06-12

QuBE/Qubex: an integrated hardware-software system for superconducting qubit experiments with broadband control

arXiv:2606.13010v1 Announce Type: new Abstract: Achieving high-fidelity operation in large-scale superconducting qubit systems requires not only control hardware with broad frequency coverage, low crosstalk, and tight synchronization but also software that coordinates system configuration, experiment execution, and data analysis. Here we present an integrated qubit-control system that combines broadband microwave hardware with a pulse-level software stack for scalable superconducting qubit experiments. The hardware provides broadband microwave coverage, including an instantaneous span of up to 1.6 GHz from a control output, while the software reduces setup and calibration overhead through automated configuration and built-in experiment workflows. We validate the system on a 64-qubit fixed-frequency transmon chip through full-chip frequency identification and representative demonstrations, including multi-unit far-detuned cross-resonance calibration and benchmarking that yields a measured two-qubit gate fidelity of 98.34%, and multilevel readout beyond the computational subspace. By disclosing the hardware architecture and releasing the software stack as open source, this work provides an inspectable hardware-software foundation for scalable superconducting qubit control experiments.

12.
arXiv (CS.AI) 2026-06-16

FragFuse: Bypassing Access Control of Large Language Model Agents via Memory-Based Query Fragmentation and Fusion

arXiv:2606.15609v1 Announce Type: cross Abstract: Large language model (LLM) agents increasingly rely on long-term memory to support complex task execution, user personalization, and domain adaptation. Meanwhile, emerging access-control mechanisms for LLM agents are being explored to block policy-violating requests and prevent misuse. We reveal a novel attack surface arising from agent memory operations: prohibited content that would trigger access control can be fragmented across interactions, stored in long-term memory in benign-appearing form, and later reconstructed through memory retrieval without appearing explicitly in the final user query. We propose FragFuse, the first attack that enables unprivileged users to bypass agent access control by exploiting this temporal channel introduced by long-term memory. FragFuse operates in three stages: (1) identifying rejection-responsive fragments via black-box adaptive querying with fragment masking; (2) injecting these fragments into memory using marker carrier queries; and (3) retrieving and fusing the stored fragments through a follow-up attack query. Although FragFuse can be instantiated manually for individual agents, we further develop a surrogate-based optimization scheme that tunes fusion instructions and marker designs, enabling automated attack generation without violating the attacker's threat-model assumptions. We evaluate FragFuse across four representative agent settings and task domains, covering three state-of-the-art agent access-control mechanisms. FragFuse achieves an average bypass success rate of 86.3% and an average end-to-end harmful task success rate of 41.1% across all settings, with only 4.4% average task-success degradation compared with configurations without access control. We also show that alternative defenses, including state-of-the-art prompt-injection detectors and perplexity detectors, do not effectively address this attack.

13.
arXiv (CS.LG) 2026-06-11

A theory of learning data statistics in diffusion models, from easy to hard

arXiv:2603.12901v2 Announce Type: replace-cross Abstract: While diffusion models have emerged as a powerful class of generative models, their learning dynamics remain poorly understood. We address this issue first by empirically showing that standard diffusion models trained on natural images exhibit a distributional simplicity bias, learning simple, pair-wise input statistics before specializing to higher-order correlations. We reproduce this behaviour in simple denoisers trained on a minimal data model, the mixed cumulant model, where we precisely control both pair-wise and higher-order correlations of the inputs. We identify a scalar invariant of the model that governs the sample complexity of learning pair-wise and higher-order correlations that we call the diffusion information exponent, in analogy to related invariants in different learning paradigms. Using this invariant, we prove that the denoiser learns simple, pair-wise statistics of the inputs at linear sample complexity, while more complex higher-order statistics, such as the fourth cumulant, require at least cubic sample complexity. We also prove that the sample complexity of learning the fourth cumulant is linear if pair-wise and higher-order statistics share a correlated latent structure. Our work describes a key mechanism for how diffusion models can learn distributions of increasing complexity.

14.
arXiv (CS.CL) 2026-06-11

LLMpedia: A Transparent Framework to Materialize an LLM's Encyclopedic Knowledge at Scale

Benchmarks like MMLU suggest flagship language models approach factuality saturation above 90\%. LLMpedia shows this picture is incomplete. We materialize ${\sim}$1.3M encyclopedia articles entirely from parametric memory across three model families, then audit every claim against Wikipedia and curated web evidence. For \texttt{gpt-5-mini}, the verifiable true rate is 68.4\% on Wikipedia-covered subjects - more than 21\,pp below MMLU - and the gap is driven by unverifiability (30.5\%), not refutation (1.2\%). Beyond Wikipedia, frontier articles audited against curated web evidence reach 57.6\%; Wikipedia covers only 56.7\% of model-surfaced subjects, and three model families overlap in just 7.3\% of subject choices. In a retrieval-trap benchmark inspired by prior analysis of Grokipedia, LLMpedia is more factual at roughly half the textual similarity to Wikipedia. Every prompt, article, and verdict is released. Data, code, interface: https://llmpedia.net.

15.
arXiv (CS.CV) 2026-06-16

Stepwise Token Selection for Efficient Multimodal Large Language Models

In multimodal large language models (MLLMs), inference cost is largely dominated by the visual token prefix rather than the language backbone, making token reduction a key factor for improving efficiency. Existing approaches typically assign independent importance scores to visual tokens and retain a fixed number of top-ranked tokens, implicitly assuming token independence and a uniform compression ratio across inputs. In this work, we reformulate visual token pruning as a sequential decision-making process. Specifically, we introduce a pointer-style selection mechanism that iteratively chooses informative tokens, conditioning each decision on previously selected ones, and dynamically determines when to stop via a learned termination action. This enables joint optimization of both the selected subset and its size. To enable end-to-end training under standard language modeling objectives, we design a differentiable relaxation based on a variance-preserving noise interpolation scheme, allowing gradients to propagate through the discrete selection process. Extensive experiments on LLaVA-v1.5-7B and Qwen2.5-VL-7B demonstrate that our approach consistently outperforms fixed-ratio baselines across different compression levels. Under aggressive pruning that removes 88.9% of visual tokens, our method preserves 94.6% of the original accuracy while achieving a 1.88x speed-up in prefill latency.

16.
arXiv (quant-ph) 2026-06-16

Analyzing Initialization Strategies for the Local Unitary Cluster Jastrow Ansatz within the Quantum-Centric Supercomputing Framework

arXiv:2606.14933v1 Announce Type: cross Abstract: In this study, we analyze the choice of local unitary cluster Jastrow (LUCJ) ansatz initialization and sensitivity of the sample-based quantum diagonalization (SQD) algorithm within the quantum-centric supercomputing (QCSC) framework. We examine six initialization strategies, including those based on coupled-cluster singles and doubles (CCSD), M{\o}ller-Plesset second-order perturbation theory (MP2), data-driven coupled-cluster (DDCC), and trivial (zeroes and random) initializations, across twelve molecular systems and three basis sets (STO-3G, cc-pVDZ, and aug-cc-pVDZ). We find that while the mean absolute percentage errors (MAPEs) between the alternative and CCSD-initialized t2-amplitudes span many orders of magnitude, the resulting SQD energies are largely insensitive to this variation. In particular, most initializations recover energies within chemical accuracy (+/-1.6 mEh) of the CCSD reference, with convergence improving as the basis set size increases. Notably, random initialization achieves performance competitive with CCSD across all basis sets, while zeroes initialization, despite having smaller deviations from CCSD, yields the worst energy agreement. Our results highlight that the proximity to the CCSD initialization is not a reliable predictor of the quality of electronic energies. These findings establish that configuration recovery within SQD, rather than circuit initialization, is the dominant factor governing energy accuracy, and suggest that computationally cheaper initialization strategies are viable alternatives to CCSD for QCSC workflows

17.
arXiv (CS.AI) 2026-06-16

PISA: A Pragmatic Psych-Inspired Unified Memory System for Enhanced AI Agency

arXiv:2510.15966v2 Announce Type: replace Abstract: Memory systems are fundamental to AI agents, yet existing work often lacks adaptability to diverse tasks and overlooks the constructive and task-oriented role of AI agent memory. Drawing from Piaget's theory of cognitive development, we propose PISA, a pragmatic, psych-inspired unified memory system that addresses these limitations by treating memory as a constructive and adaptive process. To enable continuous learning and adaptability, PISA introduces a trimodal adaptation mechanism (i.e., schema updation, schema evolution, and schema creation) that preserves coherent organization while supporting flexible memory updates. Building on these schema-grounded structures, we further design a hybrid memory access architecture that seamlessly integrates symbolic reasoning with neural retrieval, significantly improving retrieval accuracy and efficiency. Our empirical evaluation, conducted on the existing LOCOMO benchmark and our newly proposed AggQA benchmark for data analysis tasks, confirms that PISA sets a new state-of-the-art by significantly enhancing adaptability and long-term knowledge retention.

18.
arXiv (CS.CL) 2026-06-16

SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents

Large language model (LLM) agents increasingly rely on agent harnesses that manage context, tools, and multi-turn execution, making tools a central interface for acting in realistic digital environments. As harness-connected tool ecosystems expand to hundreds or thousands of APIs, services, and task-specific skills, exhaustive tool schema injection becomes costly and imposes a closed-world assumption that limits agents to a predefined static inventory. Retrieval-augmented tool selection offers a natural alternative, but existing one-shot retrieval methods often fail to align isolated tool descriptions with the agent's true task intention, especially in long-horizon tasks where required capabilities emerge through decomposition, observations, and newly induced subgoals. We propose SING, an intention-aware active tool discovery framework that builds an intention-tool graph linking user intentions, tool capabilities, and tool collaboration patterns, and dynamically retrieves tools according to evolving task states. Using a unified corpus of 7,471 tools, we evaluate SING on three real-world tool-use benchmarks. SING improves Global Recall@5 by up to 59.8% and downstream success rate by up to 28.9% over baselines, while reducing full-corpus tool-schema exposure by 99.8%, demonstrating that intention-aware graph structure enables more accurate and context-efficient tool discovery in large-scale agentic ecosystems.

19.
arXiv (CS.CL) 2026-06-17

ALAS: An Automatic Latent Alignment Score for Audio Language Models

Large Language Models (LLMs) are extended into Speech-LLMs, and the quality of the audio–text alignment they learn affects most downstream Spoken Language Understanding (SLU) behavior. Yet despite a growth of fusion strategies, there is no standard way to measure how well a Speech-LLM internally binds audio frames to text tokens. We introduce ALAS (Automatic Latent Alignment Score), a model and task-agnostic metric that probes the LLM's per-layer hidden states, scoring the cross-modal cosine similarity between audio and text representations against a Whisper-derived reference. ALAS needs only a frozen forward pass and an off-the-shelf ASR reference, with no training or fitted classifier, and is calibrated to an interpretable uniform baseline comparable across tasks. Applying ALAS to four open-source Speech-LLMs (AF3, Qwen2-Audio, Qwen-Omni, SALMONN) across emotion recognition (IEMOCAP), open-ended SQA (LibriSQA), and multi-choice audio understanding (MMAU-speech), we find that the depth and strength of alignment reflect each model's audio-encoder design and the acoustic-versus-semantic demands of the task, and that ALAS tracks but does not duplicate task accuracy, exposing models that score well without genuinely grounding in the audio. We release ALAS as an open-source library so that practitioners can probe their own Speech-LLMs or try it on new tasks.

20.
arXiv (CS.AI) 2026-06-16

LiteOdyssey: A Lightweight Reasoning AI Agent for Interpretable Rare-Disease Diagnosis

arXiv:2606.16149v1 Announce Type: new Abstract: Most medical AI systems improve by scaling additional machinery: more fine-tuning data, more agents, and/or larger retrieval databases. In rare-disease diagnosis, however, such scaling can produce systems that are difficult to deploy, audit, and maintain. We asked whether state-of-the-art diagnostic performance could instead be achieved by extending the reasoning chain of a single AI agent: guiding it with a diagnostic policy, developed through human-AI collaboration and augmenting with freely available biomedical tools. We introduce LiteOdyssey, a lightweight rare-disease diagnostic framework that guides reasoning language model through a clinical genetics workflow. This framework was developed through Policy Iteration with Human Feedback (PIHF) and uses dynamic access to public biomedical tools. On two challenging benchmarks that provide only patient clinical features, LiteOdyssey achieved state-of-the-art performance, with an overall disease Recall@1 of 59.3% over the combined 1,243 cases of LIRICAL (n = 370) and the PhenoPacket Store (n = 873). Both benchmarks have a high proportion of ultra-rare disease (a prevalence below 1 in 1,000,000, with ultra-rare shares of approximately 45% and 52.8%, respectively). On the more difficult PhenoPacket subset, where causal diseases were not mapped to Orphanet in our rarity-mapping pipeline, LiteOdyssey achieved 60.7% Recall@1, compared with 10.7% for the same baseline model (GPT-5.4) without tools. This performance was achieved without fine-tuning, multi-agent ensembles, or a large case-retrieval database. Gains were also observed in the following: on cases never seen during development, on a private cohort of real-world rare disease patients, and on a smaller open-weights model. LiteOdyssey suggests a path toward rare-disease AI systems that are accurate, easier to deploy, and more transparent for physician review.

21.
arXiv (CS.CL) 2026-06-12

Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents

LLM-based agents are becoming increasingly capable, yet their safety lags behind. This creates a gap between what agents can do and should do. This gap widens as agents engage in multi-turn interactions and employ diverse tools, introducing new risks overlooked by existing benchmarks. To systematically scale safety testing into multi-turn, tool-realistic settings, we propose a principled taxonomy that transforms single-turn harmful tasks into multi-turn attack sequences. Using this taxonomy, we construct MT-AgentRisk (Multi-Turn Agent Risk Benchmark), the first benchmark to evaluate multi-turn tool-using agent safety. Our experiments reveal substantial safety degradation: the Attack Success Rate (ASR) increases by 16% on average across open and closed models in multi-turn settings. To close this gap, we propose ToolShield, a training-free, tool-agnostic, self-exploration defense: when encountering a new tool, the agent autonomously generates test cases, executes them to observe downstream effects, and distills safety experiences for deployment. Experiments show that ToolShield effectively reduces ASR by 30% on average in multi-turn interactions. Our code is available at https://github.com/CHATS-lab/ToolShield.

22.
arXiv (CS.CL) 2026-06-11

Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code

Large Language Models (LLMs) are increasingly used for code generation, raising concerns that they may be misused to produce malicious code. Meanwhile, Grammar-Constrained Decoding (GCD) has been widely adopted to improve the reliability of LLM-generated code by enforcing syntactic validity. In this paper, we reveal a counterintuitive risk: this reliability-oriented technique can itself become an attack surface. We uncover a new jailbreak attack, termed CodeSpear, that exploits GCD to induce LLMs into generating malicious code. Our experiments show that simply applying a benign code grammar constraint can effectively jailbreak LLMs. To address this vulnerability, we propose CodeShield, a safety alignment approach that robustly preserves safe behavior even under attacker-controlled grammar constraints. CodeShield aligns the model in the code modality by teaching it to generate honeypot code under GCD. Such code is semantically harmless, so it does not implement the malicious request, and structurally diverse, so it is difficult to suppress through grammar tightening. At the same time, CodeShield still preserves natural-language refusals when natural language is available. Experiments on 10 popular LLMs across 4 benchmarks show that CodeSpear outperforms representative jailbreak baselines and increases the attack success rate by more than 30 percentage points on average. CodeShield also restores safety under CodeSpear while preserving benign utility. Our findings reveal a fundamental risk of GCD and call for greater attention to its potential security implications.

23.
arXiv (CS.CL) 2026-06-15

Creative Integration: A Decidable Criterion of Creativity

"Integrative" solutions are widely praised but rarely defined: we lack an operational way to tell a genuine integration – one that makes the world cheaper to describe – from a tidy re-description. Building on the lineage that treats creativity and intelligence as compression, we give such a criterion for creative integration (CI): the resolution of a real conflict between A and B is CI if and only if, under a fixed description language, the description length strictly shrinks (C = L_pre/L_post > 1), with the reduction located in the conflict itself. We make the judgment decidable through four binary, conjunctive gates, and we fix its extension through a taxonomy of pseudo-integration that names and rejects the look-alikes. We back the criterion with a curated, multi-domain corpus and – crucially – validate it not by human inter-rater agreement but by four falsifiable tests it could fail: an independent computational check, discrimination against hard negatives, out-of-sample prediction, and description-language robustness; all pass with margin. The contribution is not "creativity is compression" but its decidability, discrimination, and corpus: on this account, what makes a move genuinely creative – rather than merely novel – is that it compresses a conflict, with novelty and value as downstream symptoms; whether all creativity is so constituted we state as an explicit conjecture. We claim only the sign of C-1; we judge, not generate. The result is a citable primitive for a broader program.

24.
medRxiv (Medicine) 2026-06-18

Factor Analysing Predictive Processing: No Evidence for a General Factor Across Tasks

Background & Hypothesis: Dysfunctional predictive processing (PP), specifically the aberrant weighting of priors, is a frequently-proposed mechanism for psychosis and psychosis-like phenomena (schizotypy). Evidence for this theory mostly originates from single-task studies, which assume that all tasks load onto a single latent construct of PP performance, but the underlying factor structure of PP tasks is unknown. PP deficits in psychosis may be better described by a two-factor, hierarchical model: weakened lower-level (perceptual) priors compensated by higher-level (cognitive) priors. Study Design: This study implements a multi-paradigm approach in healthy participants to investigate latent constructs underlying PP and their relationship to schizotypy. Participants (N = 73) completed 6 tasks measuring reliance on priors across language, memory, visual, and auditory domains. A factor analysis investigated whether performance across tasks is captured by a single or two-factor model. Study Results: Although a two-factor model best described performance, factors reflected within-task correlations rather than a PP hierarchy. Cross-task PP measures were poorly correlated, suggesting that individuals' weighting of priors was task-specific. A full model including all task outcomes (not factors) significantly predicted the severity of schizotypal aberrant beliefs but no other schizotypal measures. Conclusions: These results do not evidence a single factor underpinning PP performance. It is therefore inappropriate to use results from single tasks to propose a generalised PP deficit in psychosis. Variation was also not captured by a two-factor hierarchical model of priors. Further multi-paradigm research is required to evaluate alternative models or additional variables that describe aberrant PP in psychosis.

25.
arXiv (CS.LG) 2026-06-19

Critical Percolation as a Synthetic Data Model for Interpretability

arXiv:2606.20347v1 Announce Type: new Abstract: Neural networks learn features that reflect the hierarchical, multi-scale structure of natural data. Synthetic datasets used to evaluate interpretability methods typically lack this structure, limiting their value as realistic toy models. To close this gap, we introduce a family of synthetic datasets consisting of hierarchical functions defined on critical mean-field percolation clusters embedded in a high-dimensional data space. The percolation data consists of sparse, low-dimensional fractal clusters with a power-law size distribution. Latent variables modeling a taxonomic hierarchy generate each data point's target value. The data model is analytically tractable with known critical exponents that fix its properties without requiring hyperparameter tuning. We leverage a mapping between percolation clusters, random trees, and additive coalescence to propose an almost linear-time algorithm to jointly sample a random tree and its hierarchical latent decomposition, enabling data generation at arbitrary scale. Using probing experiments, we find that the model's ground-truth latent variables can be linearly decoded from neural network activations. Together, sparsity, self-similarity, power-law statistics, and analytical tractability make critical percolation a principled testbed for interpretability research.