cs.AI · 2026-06-23 · No. 32

Artificial Intelligence, 2026-06-23.

38 new papers in cs.AI. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

38 entries
  1. 01

    PsyBridge: A Hybrid Intelligent Framework for Multi-Dimensional Mental Health Assessment and Decision Support

    Sunil Wanjari, Manish Thakre, Aayushi Asole, Sharwari Raut, Kwabena Adu-Duodu, Yinhao Li, Stanly Wilson

    cs.AI · cs.LG

    Mental health assessment commonly relies on isolated screening instruments or data-driven models that often lack interpretability and multi-dimensional integration. Existing approaches frequently focus on individual indicators such as depression or anxiety while providing limited support for comprehensive and explainable decision-making. To address this limitation, this study proposes PsyBridge, a hybrid intelligent decision-support framework...

    arxiv.org/abs/2606.23673 · PDF

  2. 02

    Teaching LLMs String Matching, Backtracking, and Error Recovery to Deduce Bases and Truth Tables for the Combinatorially Exploding Bit Manipulation Puzzles

    Prateek Agnihotri, Sanchit Jain, Prabhat Agnihotri, Aditya Prasad, Shubham Jain

    cs.AI

    This paper presents our algorithmic innovations for the NVIDIA Nemotron Model Reasoning Challenge, focusing on Bit Manipulation Puzzles. In this task, the objective is to discover a hidden logical rule transforming input binary strings to outputs, then apply it to unseen inputs. Large Language Models (LLMs) notoriously struggle here; traditional methods force them to simulate complex boolean logic and arithmetic, leading to hallucinations....

    arxiv.org/abs/2606.23672 · PDF

  3. 03

    TailorMind: Towards Preference-Aligned Multimodal Content Generation

    Hengji Zhou, Ye Liu, Yufeng Liu, Si Wu, Lianghao Xia, Liqiang Nie

    cs.AI

    Personalized content systems depend on available UGC and struggle when suitable content is absent, delayed, or costly to create. Although multimodal generators can synthesize content on demand, how to translate behavioral traces into generation-ready preferences remains underexplored. We study personalized multimodal content generation: creating user-tailored multimodal content without existing item pools or waiting for matching UGC. We...

    arxiv.org/abs/2606.23643 · PDF

  4. 04

    AI Exposure Scores: what they measure, what they miss, and what comes next

    Campbell Lund, Thomas Euyang, Zanele Munyikwa, Marzieh Fadaee

    cs.AI · econ.GN

    A set of exposure scores calculated in 2023 has become a central empirical input to the future of work debate. Produced by Eloundou et al. (2023) and referred to here as the GPTs are GPTs scores, they define exposure as the share of occupational tasks a large language model can assist with. This work is a genuine methodological contribution, but as the scores travel from the time and place they were produced, the limitations the authors named...

    arxiv.org/abs/2606.23633 · PDF

  5. 05

    AI-driven Optimisation of Quality of Recovery (QoR) in Remote Patient Monitoring

    Yansong Liu, Li-Hsi, Lin, Pramit Khetrapal, Ronnie Stafford, John Kelly, Ivana Drobnjak

    cs.AI

    Remote patient monitoring depends on patient-reported data to capture the subjective dimension of recovery that devices cannot measure. The Quality of Recovery (QoR-15) survey is the gold-standard instrument for this purpose. It was designed and validated for occasional in-hospital assessment, yet remote monitoring now administers it to patients daily. In our own post-surgical deployment, only 55% of patients submitted the survey more than 14...

    arxiv.org/abs/2606.23631 · PDF

  6. 06

    Causal Discovery in the Era of Agents

    Yujia Zheng, Vishal Verma, Mantej Gill, Haoyue Dai, Peter Spirtes, Kun Zhang

    cs.AI · cs.LG · cs.SE · stat.AP

    Recent attempts to combine large language models (LLMs) with causal discovery ask models to infer pairwise directions, propose graph structures, or inject language-model outputs as priors and constraints. These approaches promise faster analysis, but they also obscure whether a causal evidence is supported by data and assumptions or by textual associations, prompt artifacts and hallucinated mechanisms. We argue for a different role for agents...

    arxiv.org/abs/2606.23608 · PDF

  7. 07

    Against Proxy Optimization

    Sven Neth

    cs.AI

    I discuss conditions under which maximizing a proxy utility function is harmful and suggest this poses problems for applying decision theory.

    arxiv.org/abs/2606.23597 · PDF

  8. 08

    SPIRAL: Learning to Search and Aggregate

    Jubayer Ibn Hamid, Ifdita Hasan Orney, Michael Y. Li, Omar Shaikh, Yoonho Lee, Dorsa Sadigh, Chelsea Finn, Noah Goodman

    cs.AI

    Language model reasoning can be substantially improved at test time via scaffolds that scale inference compute across different primitives -- sequential reasoning within a trace, independently sampled parallel traces, and aggregation of multiple reasoning traces into a final response. During post-training, however, language models are optimized only for sequential reasoning within a single trace. We introduce Sequential-Parallel-Aggregative...

    arxiv.org/abs/2606.23595 · PDF

  9. 09

    The Topology of Ill-Posed Questions: Persistent Homology for Detection and Steering in LLMs

    Guangyu Jiang, Sizhe Tang, Mahdi Imani, Tian Lan

    cs.AI

    Ill-posed questions, including ambiguous, underspecified, or contradictory queries, may admit no valid answer or multiple plausible answers, posing a challenge for large language models (LLMs). Existing approaches largely analyze ill-posedness through model outputs and often focus on specific subclasses. We investigate whether diverse sources of ill-posedness can be represented within a unified topology of LLM internal states and whether this...

    arxiv.org/abs/2606.23590 · PDF

  10. 10

    VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct

    Haoling Li, Kai Zheng, Jie Wu, Can Xu, Qingfeng Sun, Han Hu, Yujiu Yang

    cs.AI · cs.CL · cs.CV · cs.LG

    Scaling reinforcement learning for visual mathematical reasoning requires more than generating harder questions: as data volume grows, the reward labels themselves must remain reliable. Yet existing data pipelines scale supervision while trusting the labeller, and policy-side methods assume the underlying answers are already correct. We instead treat scaling as a verifiable data-construction problem and decouple two axes before any policy...

    arxiv.org/abs/2606.23543 · PDF

  11. 11

    POTracker: Optimizing Large Language Models for Standard-Compliant Power Outage Report Generation

    Hung Phan, Aniroop Naladala, Dubey Avanindra, Supryia Chinthavali, Lunga Dalton, Ali Jannesari

    cs.AI

    Recent large language models (LLMs) are good at general text generation, but it is still hard to use them for domain-specific data generation because the output must follow strict formatting and structural rules. Unlike open-ended tasks such as question answering or translation, domain-specific generation must be both semantically correct and compliant with existing guidelines and standards. In this work, we study the nationwide...

    arxiv.org/abs/2606.23533 · PDF

  12. 12

    CADRE: Stable, Parameter Efficient Adaptation of Medical Vision Language Models with Bounded Forgetting and Prior Drift

    Amrita Singh, Rishabh Jha

    cs.AI

    Medical vision-language models (VLMs) such as BiomedCLIP generalize broadly, but adapting them to a clinical service is as much a safety problem as an accuracy one. Updating a deployed model for a new imaging modality can fail silently in two ways that harm patients: it can forget modalities it already handled (catastrophic forgetting), and it can drift from its trustworthy pretrained prior toward modality-specific shortcuts. We study...

    arxiv.org/abs/2606.23487 · PDF

  13. 13

    AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction

    Shanhui Zhao, Jiacheng Liu, Guohong Liu, Jichao Yan, Jialei Ye, Yuhao Yang, Hao Wen, Shizuo Tian, Yizhen Yuan,...

    cs.AI · cs.OS

    AI agents are driving a new software paradigm, with the ability to autonomously call tools, extract information, manage memory, and complete tasks that span applications and data sources. Most existing end-user operating systems, however, are designed for application-centric workflows and offer little native support for AI agents. This mismatch limits the wider adoption of agents and leads to execution overhead and safety risks when running...

    arxiv.org/abs/2606.23449 · PDF

  14. 14

    Cross-Architectural Mixture-of-Experts with Adaptive Soft Routing for Plant Leaf Disease Classification

    Phi-Hung Hoang, Thi-Thu-Hong Phan

    cs.AI

    Plant leaf disease classification is crucial for crop protection and precision agriculture but remains challenging under complex backgrounds, illumination variations, and severe class imbalance. Moreover, single-architecture models often fail to effectively capture both local and global representations. To address these challenges, this study proposes an adaptive soft Mixture-of-Experts (MoE) framework with cross-architectural routing that...

    arxiv.org/abs/2606.23441 · PDF

  15. 15

    Digital Humanism and Evolutionary Design

    Wolfgang Höhl

    cs.AI · cs.CY

    This paper examines the two concepts of digital humanism and evolutionary design. The aim is to identify and highlight potential common structures, synergies, and challenges. How should and can technical systems be designed, and what implications does this have for the design of our environment? In light of the current debate surrounding artificial intelligence, this paper aims to serve as a preliminary study to help better understand the two...

    arxiv.org/abs/2606.23417 · PDF

  16. 16

    Litmus: Zero-Label, Code-Driven Metric Specification for Evaluating AI Systems

    Prajjwal Gupta, Prasang Gupta, Vishal Bhutani, Apoorva Sharma, Sumanth Chundru, Waqar Sarguroh, Kevin Paul

    cs.AI

    As agentic LLM systems move from prototypes to deployment across increasingly diverse domains, evaluating them has become both more important and more difficult. The challenge is not only that individual metrics may be unreliable, but that evaluation goals are often left implicit. Without a clear account of what a system is expected to do, how it can fail, and which failures matter, metric choices become difficult to justify, interpret, or...

    arxiv.org/abs/2606.23403 · PDF

  17. 17

    Abstract representational geometry supports inference in large language models

    Yunan Zeng, Yuwang Wang

    cs.AI

    A defining feature of human intelligence is the ability to adapt to changing environments by inferring latent task structure from sparse observations. Neuroscientific research indicates that this capability relies on the hippocampus constructing abstract representations, expressed as low-dimensional, approximately orthogonal manifolds in neural state space. However, the internal mechanisms of large language models (LLMs) remain largely...

    arxiv.org/abs/2606.23345 · PDF

  18. 18

    EHR-Complex: Benchmarking Medical Agents for Complex Clinical Reasoning

    Yitong Qiao, Lei Liu, Yue Shen, Jian Wang, Jinjie Gu, Zhixuan Chu, Kui Ren

    cs.AI

    Clinical agents promise to democratize access to electronic health records (EHRs), yet existing benchmarks fail to reflect the complexity of practical EHR analysis, e.g., often operating on idealized, clean EHRs via static SQL generation rather than interactive execution. In this work, we introduce EHR-Complex, a large-scale benchmark designed for interactive clinical database reasoning. Built on the large MIMIC-IV substrate (365K patients,...

    arxiv.org/abs/2606.23301 · PDF

  19. 19

    GIF: Locally Sound Geometric Information Flow Control for LLMs

    Adam Storek, Nikolaus Holzer, Zhuo Zhang, Suman Jana

    cs.AI

    Large language models increasingly mediate interactions between sensitive data, untrusted inputs, and privileged actions in agentic systems, creating security and privacy risks. These range from prompt injections that manipulate downstream tool use to leakage of confidential information through model outputs. Recent Information Flow Control (IFC)-based defenses show promise but lack a principled semantic foundation for reasoning about...

    arxiv.org/abs/2606.23277 · PDF

  20. 20

    HOLMES: Evaluating Higher-Order Logical Reasoning in LLMs

    Yucheng Wu, Jundong Xu, Mingzhen Ju, Yue Yu, Chenpeng Wang, Haoxuan Li, Liangming Pan

    cs.AI

    Logical reasoning is essential for reliable AI, yet existing benchmarks are largely first-order-logic-centric, focusing on object-level deduction over fixed predicates. This misses many realistic scenarios where models must reason over rules, predicates, functions, constraints, and decision procedures themselves. We introduce HOLMES (Higher-Order Logic Meets real-world Explainable Symbolic reasoning), the first real-world benchmark for...

    arxiv.org/abs/2606.23238 · PDF

  21. 21

    SPADE: Structure-Prior Adaptive Decision Estimation

    Yifan Wang

    cs.AI

    Physical-structure priors such as conservation laws, Hamiltonian forms, and symmetries can improve scientific machine learning when correct, but can degrade predictions when misspecified. Existing methods usually enforce a chosen structure or tune a soft penalty, without a calibrated rule for deciding whether to impose a prior, how strongly to impose it, which prior to use, or which subset of candidate laws holds. We introduce SPADE,...

    arxiv.org/abs/2606.23219 · PDF

  22. 22

    Capable but Careless: Do Computer-Use Agents Follow Contextual Integrity?

    Anmol Goel, Iryna Gurevych

    cs.AI · cs.CL

    Computer-use agents (CUAs) now act on a user's behalf across personal applications such as email, calendars, and to-do lists. This cross-application access is useful, but it also creates a privacy risk that has been largely overlooked: when an agent works in one context, it can pull in information from another that is inappropriate in that context. Hence, we introduce AgentCIBench, an evaluation harness that turns this risk into executable,...

    arxiv.org/abs/2606.23189 · PDF

  23. 23

    DART: Draft-Agreement Routing for Training-Free Adaptive Thinking Budgets in Hybrid Reasoning Models

    Jungseob Lee, Seongtae Hong, Seungjun Lee, Jaehyung Seo, Junyoung Son, Sugyeong Eo, Chanjun Park, Hyeongju Park,...

    cs.AI · cs.CL

    Hybrid reasoning models can answer directly or spend extra tokens on extended thinking. A practical router should choose between these modes for each query, so easy problems avoid unnecessary reasoning and hard problems receive enough budget to finish the answer. Existing routers move in this direction, but they typically require labeled training data or fix thinking budgets up front, ignoring answer-level evidence from the model itself. We...

    arxiv.org/abs/2606.23181 · PDF

  24. 24

    Decomposing Financial Market Dynamics via Mechanism Analysis in an Evolutionary Multi-Agent Simulation

    Zhibao Chen

    cs.AI · cs.MA · cs.NE

    Evolutionary agent-based markets (ABMs) couple several mechanisms -- who reproduces, how price forms, how biased the agents are, how consensus propagates -- yet these are usually fixed by convention, so it is unclear which mechanism controls which emergent property. In a coevolving, endogenous-price simulator with 120 heterogeneous behavioral agents, we make four mechanisms pluggable and run matched 3x20-seed interventions. We find the levers...

    arxiv.org/abs/2606.23158 · PDF

  25. 25

    Managing Procedural Memory in LLM Agents: Control, Adaptation, and Evaluation

    Julia Belikova, Rauf Parchiev, Evgeny Egorov, Grigorii Davydenko, Gleb Gusev, Andrey Savchenko, Maksim Makarenko

    cs.AI · cs.CL · cs.SE

    Procedural memory is increasingly used to improve LLM agents on recurring workplace tasks, yet its ability to produce reusable skills remains poorly understood. We introduce AFTER, a benchmark of 382 realistic enterprise tasks spanning six professional roles and 22 procedural skills, designed to evaluate how skills transfer across tasks, roles, and model backbones. The benchmark includes controlled evaluation settings for local improvement,...

    arxiv.org/abs/2606.23127 · PDF

  26. 26

    A Matter of Time: Towards a General Theory of Agency

    Amahury J. López-Díaz, Carlos Gershenson

    cs.AI · q-bio.OT

    Agency is often invoked in research on philosophy, biology, and cognitive science without a clear account of how it originates from material organization. Building on temporally parametrized (F, A)-systems, this paper develops a graded organizational theory of agency grounded in relational biology, physical biosemiotics, and process ontology. We argue that self-referential closure cannot be adequately conceived outside time: once the...

    arxiv.org/abs/2606.23122 · PDF

  27. 27

    TTFT-Aware Graph Chain-of-Thought:Distance-Indexed Neural A* for Low-Hallucination Multi-Hop Medical Reasoning

    Bechir Dardouri, Kaïs Zhioua, Yassine Msaddak

    cs.AI

    Hallucinations and opaque reasoning remain unacceptable failure modes for clinical LLMs. We present a production-grade GraphRAG stack that constrains answers to verifiable graph chain-of-thought paths in a heterogeneous, ~700K-node medical knowledge graph powering a fertility assistant. The core idea is targeted navigation: a directed Pruned Landmark Labeling (PLL) oracle provides exact distances for sub-millisecond feasibility checks and...

    arxiv.org/abs/2606.23108 · PDF

  28. 28

    Cognitive Digital Twins: Ethical Risks and Governance for AI Systems That Model the Mind

    Vamshi Krishna Bonagiri, Juan Nicolas Sepulveda-Arias, Abdoul Jalil Djiberou Mahamadou, Monojit Choudhury

    cs.AI · cs.CL

    As AI systems become increasingly persistent and personalized, they make possible a class of technologies that we call cognitive digital twins (CDTs): dynamic computational representations of a specific person's cognition, updated from behavioral, contextual, or physiological data in order to model, predict, or simulate that person's cognition, or to act as that person's communicative or decision-making proxy. CDTs combine cognitive inference...

    arxiv.org/abs/2606.23094 · PDF

  29. 29

    Some Results about the Expressivity of Preference-Incomplete Structured Argumentation Frameworks

    Antonio Yuste-Ginel

    cs.AI · cs.LO

    This paper studies the expressive power of ASPIC$^+$ argumentation frameworks with uncertain preference profiles by comparing them with several abstract formalisms with uncertain defeats. Most of our results are negative (and some of them are theoretically unexpected). We also conjecture a positive, non-trivial threshold for the expressivity of uncertain preferences, and prove some essential preliminary steps toward the confirmation of this...

    arxiv.org/abs/2606.23055 · PDF

  30. 30

    IPO Finance Agent: Evaluation of LLM Financial Analysts beyond Finance Agent v2, with Automated Rubric Generation -- the Case of the SpaceX (SPCX) IPO

    Mostapha Benhenda

    cs.AI · q-fin.GN

    Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks. However, it narrowly deals with periodic reporting from publicly traded companies (SEC 10-K and 10-Q filings), and its agentic harness relies on naive, unenriched chunk retrieval. Neither the task design nor the retrieval approach addresses the distinct challenges of IPO due...

    arxiv.org/abs/2606.23032 · PDF

  31. 31

    From numerical proportions to analogical proportions between probabilities

    Henri Prade, Gilles Richard

    cs.AI

    Analogical proportions link four items a, b, c, d by a relation stating that ``a is to b as c is to d", a, b, c, d being the formal representation of real world entities, ranging from simple numerical values to more complex structures such as profiles. Accordingly, $a, b, c, d$ could be atomic values like Boolean, nominal or numerical values, more generally vectors of such values, or even families of items represented by logical formulas. In...

    arxiv.org/abs/2606.23029 · PDF

  32. 32

    A Stackelberg Framework for Resource-Aware LLM Agents: Learning, Repair, and Conditional Guarantees

    Baoxun Wang

    cs.AI

    Large language model (LLM) agents increasingly operate as multi-turn systems that must allocate context, prompt verbosity, and tool access under finite computational budgets. Static thresholds are simple, but they are brittle under heterogeneous tasks and evolving session states. We formulate resource governance as a contextual Stackelberg game: a controller commits to a quality target and a cost incentive, while an executor responds with...

    arxiv.org/abs/2606.23026 · PDF

  33. 33

    Joint Air Traffic Flow and Capacity Management via Answer Set Programming

    Alexander Beiser, Markus Hecher, Nysret Musliu, Stefan Woltran

    cs.AI

    Operational Air Traffic Flow and Capacity Management (ATFCM) balances flight demand with available sector capacity, to ensure safe and efficient operations. Mathematical models enhance operational ATFCM performance by framing demand-capacity balancing as an optimization problem, maximizing efficiency while adhering to safety constraints. However, SOTA research optimizes the aircraft trajectories (called ATFM) or the sector configuration...

    arxiv.org/abs/2606.22978 · PDF

  34. 34

    When Preferences Fail to Become Incentives: A Utility-Behavior Gap in Large Language Models

    Yujun Zhou, Christopher M. Ackerman

    cs.AI

    Recent work on preference elicitation in large language models (LLMs) has demonstrated that, when given a series of choices between two outcomes, LLMs reveal a coherent, model-specific utility structure. Notably, this structure often includes preferences that the models' trainers did not intend, such as valuing people of some nationalities above others, raising the possibility that LLMs might be forming emergent, misaligned goals, which, if...

    arxiv.org/abs/2606.22974 · PDF

  35. 35

    The Impact of VAE Design on Latent Pose Representations for Diffusion-based Sign Language Production

    Guilhem Fauré, Mostafa Sadeghi, Sam Bigeard, Slim Ouni

    cs.AI · cs.CV

    Latent diffusion approaches to sign language production (SLP) rely on an initial stage that learns an encoding of sign pose sequences, enabling generative modeling in the resulting latent space. The autoencoder used in this stage is typically evaluated in terms of reconstruction quality using geometric metrics common in SLP. While informative, these metrics do not fully capture latent space properties that may influence the training and...

    arxiv.org/abs/2606.22959 · PDF

  36. 36

    Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents

    Aman Mehta, Anupam Datta

    cs.AI · cs.CL

    Long-horizon agents depend on context management: systems compress, summarize, and evict old tokens so tasks can continue beyond finite windows. That is safe only when dropped information is no longer needed or has been internalized. Plans are the stress case: they are written early, used for many steps, and first to be evicted. We introduce replay pairing, a diagnostic that runs the same trajectory with and without the plan in history and...

    arxiv.org/abs/2606.22953 · PDF

  37. 37

    ENVS: Environment-Native Verified Search for Long-Horizon GUI Agents

    Yincheng Zhou, Athena Zhuoming Zhong, Shijie Zhang, Kevin Zhang, Teresa Xiaotao Shang, Shanghang Zhang

    cs.AI · cs.CV

    As multimodal agents move from interface understanding to real software control, successful trajectory discovery in live desktop environments becomes a key challenge. GUI tasks require long-horizon sequences of precise mouse and keyboard actions, while feedback is sparse, delayed, and costly to obtain through VM rollouts. We propose Environment-Native Verified Search (ENVS), a training-time search-and-filter pipeline that uses the environment...

    arxiv.org/abs/2606.22948 · PDF

  38. 38

    When Agents Commit Too Soon: Diagnosing Premature Commitment in LLM Agents

    Aman Mehta

    cs.AI

    Long-horizon LLM agents can fail quietly: they settle on one reading of the evidence early, then spend the rest of the run defending it. We call this premature commitment. Final-answer scoring misses the failure mode because it sees only the answer, not whether the process has already collapsed to a stable path. We define representational commitment as cross-run hidden-state convergence at a fixed reasoning step, and use it as an early...

    arxiv.org/abs/2606.22936 · PDF

This edition is part of The Daily Abstract — cs.AI archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.