Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

01.
arXiv (quant-ph) 2026-06-12

Efficient certification of intractable quantum states with few Pauli measurements

arXiv:2511.07300v2 Announce Type: replace Abstract: Efficient verification of quantum computational resources is crucial as experiments advance toward fault-tolerance. Universal quantum computation can be achieved by consuming resource states through simple Pauli measurements, yet a significant gap remains between states that are easy to certify and those required for universality. We focus on Clifford-enhanced Product States, a class of resource states obtained by applying Clifford circuits to a product of single-qubit, potentially magic, states. While essential for universal computation, the certification of such states has previously relied on query oracles that are \#P-hard to implement, leaving their efficient, oracle-free verification an open challenge. In this work, we demonstrate that such classically intractable resource states can be efficiently verified using only Pauli measurements. Our protocol achieves sample- and time-efficiency in both i.i.d.\ and adversarial settings. This work fills a gap in Pauli-based certification, providing a new practical pathway to verify resource states that drive universal Pauli-based quantum computation.

02.
arXiv (CS.LG) 2026-06-12

Estimating Individualized Treatment Effects in Acute Ischemic Stroke with Causal Transformation Models (TRAM-DAG): A Multi-Centre Observational Study with External RCT Validation

arXiv:2606.12623v1 Announce Type: cross Abstract: Personalized medicine in acute ischemic stroke requires moving beyond average treatment effects (ATE) to individualized treatment effect (ITE) estimates to support treatment decisions. In acute ischemic stroke, mechanical thrombectomy has been shown to be more effective on average than lysis in randomized controlled trials (RCTs), such as the MR CLEAN study. We aim to identify which individual patients benefit most from mechanical thrombectomy compared to lysis. The outcome of interest is the modified Rankin Scale (mRS) at three months, an ordinal measure of functional disability (0: no symptoms, 6: death). We demonstrate that causal transformation models on directed acyclic graphs (TRAM-DAG) can be used for ITE estimation after being fitted on observational MAGIC multi-center stroke patient data. To ensure comparability with the MR CLEAN population, which we use for validation, we train the TRAM-DAG on a MAGIC sub-population with NIHSS at admission >= 6, corresponding to one inclusion criterion of MR CLEAN. The fitted model is then used to estimate ITEs for stroke patients in the MR CLEAN population. While these ITE estimates cannot be confirmed experimentally, we show that their average is consistent with the trial's reported ATE. Furthermore, the ITE estimates correctly rank trial patients by their observed frequency of a good outcome (mRS at three months

03.
arXiv (CS.CL) 2026-06-17

Evaluating Large Language Models Abilities for Addressee, Turn-change, and Next Speaker Prediction in Meetings

We investigate turn-taking in multimodal multi-party conversations using large language models (LLMs). We construct an evaluation framework for three tasks: addressee detection, turn-change prediction, and next speaker prediction. We compare supervised models trained for these tasks, text-based LLMs, multimodal LLMs (MM-LLMs), and human subjects. Experiments on the AMI corpus showed that LLMs outperformed supervised models and humans in next speaker prediction, despite not being trained on the target domain and without access to audio or visual information. An MM-LLM performed better than text-based LLMs on addressee detection and turn-change prediction but remained below human performance, indicating difficulty leveraging raw audio-visual signals. Ablation analyses revealed that conversational context was critical, particularly for next speaker prediction. We observed that human and LLM prediction patterns were similar, and intervals with frequent turn changes were difficult for both.

04.
arXiv (CS.CV) 2026-06-12

A Multi-Modal Framework with Cross-Subject Pseudo-Labeling and Semantic Alignment for Micro-Gesture Recognition

Micro-gestures (MGs) are spontaneous and subtle body movements that frequently convey hidden human emotions. Recognizing MGs in untrimmed videos remains highly challenging due to their extremely low signal-to-noise ratio, severe long-tailed class distribution, and the inherent domain shift encountered in cross-subject evaluation scenarios. In this paper, we propose a comprehensive multi-modal framework for Track 1 of the 4th MiGA-IJCAI Challenge. To capture fine-grained representations, we design a saliency-guided multi-modal extraction pipeline integrating 68-keypoint skeleton joint coordinates, 3D heatmap volumes, and high-resolution RGB visual features. We introduce a gentle square-root smoothed weighting mechanism paired with an Orthogonal Semantic Embedding Loss to protect tail classes without compromising overall recognition capabilities. More importantly, to bridge the cross-subject generalization gap, we propose a Cross-Modal Pseudo-Labeling (CMPL) strategy for unsupervised domain adaptation, which significantly boosts single-modal robustness. A temperature-scaled soft-voting mechanism is finally utilized to alleviate overconfidence during late fusion. Extensive experiments demonstrate that our framework achieves a competitive F1-score of 68.13\%, securing the 4th place.

05.
arXiv (quant-ph) 2026-06-16

What does measuring one qubit reveal about another? $K$-networks as a directed diagnostic for quantum circuits

arXiv:2606.16549v1 Announce Type: new Abstract: Many-qubit circuit states are hard to inspect directly, so they are often summarized by pairwise graph weights. Common pairwise weights report symmetric correlations, while many circuit questions are directed and basis-specific: if qubit $i$ is measured in a given basis, how strongly does the outcome reshape the conditional state of qubit $j$? We define $K_{i\to j}$, a directed, basis-conditioned edge weight for this question. It is large when the two measurement outcomes occur with comparable probability and leave qubit $j$ in clearly different conditional states; it is zero when the source outcome is deterministic or the target states are indistinguishable. The scalar uses standard binary-ensemble distinguishability; the paper's contribution is to turn this conditional comparison into a directed network layer for circuit states. The resulting networks are computable from two-qubit reduced density matrices. They are diagnostic (not entanglement measures): for pure two-qubit states $K$ reduces to the tangle $C^2$ (squared concurrence)[WoottersConcurrence,CKWTangle], while separable mixed states can reach $K=1$. Examples on teleportation, Grover, QAOA, and random circuit families show the intended use: $K$-networks map feed-forward, phase, and interaction-graph structure that symmetric or computational-basis summaries can leave weak or absent.

07.
arXiv (CS.CL) 2026-06-16

Retrievable Gradients: Continual Post-Training Without Cumulative Weight Drift

Continual post-training enables models to absorb emerging knowledge after deployment, but repeatedly updating shared parameters can accumulate weight drift, potentially causing catastrophic forgetting and degrading general capabilities. Retrieval-augmented generation avoids such parameter drift, yet often lacks the depth of parametric knowledge integration. In this paper, we propose ReGrad (Retrievable Gradients), a new paradigm that treats gradients as retrievable units of knowledge. ReGrad pre-computes document-specific gradients offline, stores them in an indexed Gradient Bank, and retrieves only query-relevant gradients at inference time for temporary weight adaptation. However, raw language-modeling gradients are optimized for token-level document reconstruction rather than for query-driven knowledge use. We therefore introduce a bi-level meta-learning objective that reshapes document-derived gradients into generalizable adaptation signals for downstream tasks. Experiments across general and domain-specific settings show that \textsc{ReGrad} outperforms CPT and RAG baselines, enabling scalable and reversible parametric knowledge injection without accumulating weight drift.

08.
bioRxiv (Bioinfo) 2026-06-11

A multi-agent system for spine MRI report generation from multi-sequence imaging

Spinal pathology is a leading cause of pain and disability worldwide. Spine magnetic resonance imaging (MRI) is central to clinical evaluation, yet its interpretation remains complex and time-consuming, requiring integration of information across multiple imaging sequences and anatomical regions. Despite recent advances in automated MRI analysis, effectively combining multi-sequence data while preserving sequence-specific diagnostic information remains an open challenge. Here we present SpineAgent, a multi-agent framework for spine MRI report generation built upon a multi-sequence foundation model trained on routine clinical data from 32,047 patients and 453,683 MRI series, comprising a total of 13,441,191 MRI slices. To accommodate diverse modalities of sequences, we first pre-train two DINOv3-based encoders separately on T1- and T2-weighted sequences. We then introduce a continual training strategy that learns a synthesizer to embed images of other sequences using the T1 and T2 encoders, producing patient-level embedding that integrates various signals across MRI sequences. Using these embeddings, SpineAgent achieves state-of-the-art performance, with mean 10.8% AUROC improvement across 17 spinal condition-prediction tasks compared to the best competing method, and demonstrates strong generalizability under cross-manufacturer and cross-cohort evaluation. Beyond classification, SpineAgent enables pathology localization by identifying findings-relevant slices and segmenting pathological regions. It also supports multimodal image-report retrieval, providing a solid foundation for scalable and explainable MRI report generation. We further integrate these validated capabilities of SpineAgent into 37 specialized agents for condition diagnosis, pathological-region localization, and clinically-similar-cases retrieval. Finally, we incorporate their outputs as structured tokens within a Medical Report Agent trained end-to-end for report generation. Through both automated metrics and expert evaluation by five radiologists, SpineAgent achieves leading performance in spine MRI report generation. Together, SpineAgent introduces a continual training approach for multi-sequence spine MRI understanding. By decomposing report generation into clinically grounded subtasks addressed by specialized agents, the SpineAgent framework enables accurate, interpretable and generalizable spine MRI reporting across diverse imaging sequences and anatomical regions.

09.
medRxiv (Medicine) 2026-06-18

Plasma proteomics reveals clinical and mechanistic heterogeneity among individuals who develop coronary artery disease

BACKGROUND: Individuals who develop coronary artery disease (CAD) are clinically and mechanistically heterogeneous, and understanding this variation is crucial for precise risk stratification and tailored interventions. However, the molecular mechanisms that connect these two kinds of heterogeneity remain unclear, limiting progress toward biologically grounded risk stratification and targeted interventions. Here, we investigated the heterogeneity of individuals who develop CAD by leveraging plasma proteomic signatures, placed individuals along continuous metabolic gradients and revealed the molecular programs underlying these patterns, thereby linking mechanistic variation to clinical heterogeneity. METHODS AND RESULTS: From 42,803 UK Biobank participants, including 3,713 individuals who developed CAD within 10 years (incident CAD), we first identified a 320-protein panel from 2,923 baseline proteins that improved prediction of incident CAD beyond clinical risk scores. Using reverse graph embedding, we reduced the proteomic data to two dimensions and mapped each incident case onto the resulting two-dimensional latent proteomic space. These proteomic dimensions show significant associations with cardiometabolic and kidney-related clinical markers. The patterns were replicated in the EPIC-Norfolk study. Phenome-wide Cox regression analyses further linked these proteomic dimensions to 10-year incidence rates for various diseases, including type 2 diabetes, obesity, and chronic kidney disease (CKD). Furthermore, adding the proteomic dimensions to clinical variable-based Cox regression model improved prediction of 10-year incidence of CKD and other diseases, demonstrating the value of proteomic dimensions beyond conventional clinical risk factors. Moreover, individuals with prevalent CAD (diagnosed before proteomic sampling) exhibited high, metabolically adverse dimension values, indicating that these axes capture cumulative metabolic burden. Pathway enrichment analyses implicated altered extracellular matrix organization and immune programs among the proteins contributing to the proteomic dimensions. CONCLUSIONS: Our findings demonstrate that plasma proteomic signatures can dissect the heterogeneity of individuals who develop CAD in continuous phenotypic gradients, improve prediction of CAD and comorbidities, and map underlying biological mechanisms.

10.
arXiv (CS.AI) 2026-06-16

Token Reduction Should Go Beyond Efficiency in Generative Models – From Vision, Language to Multimodality

arXiv:2505.18227v4 Announce Type: replace-cross Abstract: In Transformer architectures, tokens\textemdash discrete units derived from raw data\textemdash are formed by segmenting inputs into fixed-length chunks. Each token is then mapped to an embedding, enabling parallel attention computations while preserving the input's essential information. Due to the quadratic computational complexity of transformer self-attention mechanisms, token reduction has primarily been used as an efficiency strategy. This is especially true in single vision and language domains, where it helps balance computational costs, memory usage, and inference latency. Despite these advances, this paper argues that token reduction should transcend its traditional efficiency-oriented role in the era of large generative models. Instead, we position it as a fundamental principle in generative modeling, critically influencing both model architecture and broader applications. Specifically, we contend that across vision, language, and multimodal systems, token reduction can: (i) facilitate deeper multimodal integration and alignment, (ii) mitigate "overthinking" and hallucinations, (iii) maintain coherence over long inputs, and (iv) enhance training stability, etc. We reframe token reduction as more than an efficiency measure. By doing so, we outline promising future directions, including algorithm design, reinforcement learning-guided token reduction, token optimization for in-context learning, agentic framework design, and broader ML and scientific domains.

11.
arXiv (CS.CL) 2026-06-18

Leadership as Coordination Control: Behavioral Signatures and the Recovery-Advantage Boundary in Multi-Agent LLM Teams

Authors:

Team science holds that leadership is contingent: it helps only under specific conditions, and capable, autonomous teams may need none at all. We ask the analogous question for multi-agent LLM teams: under what measurable conditions does process-level coordination control add value, and do those conditions match what team science predicts? We use behavioral signatures (majority lock-in, exploration, recovery from an incorrect round-0 consensus) and per-action ablations, clean because each controller is an explicit action set, not a monolithic prompt. We operationalize three classical leadership styles (transactional, transformational, situational) as controllers over a shared action vocabulary (explore, revise, accept, synthesize). A matched controller with the same actions but an arbitrary rule recovers no better than majority voting, so the theory-derived rule, not the vocabulary, does the work. Across four task regimes and three open-weight model families, no controller dominates by accuracy, as the contingency view predicts: transactional control matches a shared round-0 vote on all 12 (model, regime) combinations to within 1.3pp, and gains appear only on the one combination where the round-0 majority is unreliable (llama-4-scout social; situational +8pp over flat). A recovery-advantage account, tested with four boundary probes, says a controller beats plain interaction only where the round-0 majority is unreliable, the task is recoverable, and undirected interaction does not already repair it. These regions map onto contingency theory (leadership substitutes, path-goal redundancy, the situational readiness gap), so a largely null accuracy result is what the theory predicts, not a failure of the controllers. We read process-level coordination control as a contingency to be measured and theory-mapped, not a leaderboard to be topped.

12.
medRxiv (Medicine) 2026-06-11

Allostatic Load in Endometrial Cancer Disparities

Background: Endometrial cancer incidence and mortality are increasing, particularly among Black women and for aggressive subtypes. Allostatic load (AL), a composite measure of physiologic dysregulation across metabolic, cardiovascular, and immune systems, varies by racial category and tumor subtype in other cancers. Endometrial cancer is strongly associated with obesity, and it is unknown whether AL scores maintain sufficient heterogeneity to evaluate differences across subgroups or with clinical outcomes. Objective: To describe the performance of AL scoring in endometrial cancer patients and examine associations with tumor characteristics (grade/histology) and survival outcomes. Methods: We evaluated AL among 398 participants newly diagnosed with endometrial cancer. AL score was calculated by assigning 1 point for each ''high-risk'' value (by clinical reference range or distribution-based) for 15 biologic variables for vital signs, anthropometrics, blood-based biomarkers, and medical comorbidities. Results: Distribution-based thresholds for variables were used to preserve heterogeneity in this obesity-dominant context. Overall, 68.7% of Black women had high AL compared to White (56.7%), Hispanic (56.7%), and other race (32.3%) women. Decision tree analyses revealed grade-dependent associations between AL and survival. For women with low-grade tumors, higher AL was associated with poorer overall survival. For high-grade tumors, intermediate AL ([≥]4,

13.
arXiv (quant-ph) 2026-06-15

An integrated ultrahigh vacuum cluster tool for diamond surface science and single nitrogen-vacancy center measurements

arXiv:2606.13961v1 Announce Type: new Abstract: We present a custom-designed ultrahigh vacuum (UHV) cluster tool developed for studying shallow nitrogen-vacancy (NV) centers in diamond, enabling in situ diamond surface preparation, characterization, and single NV center dynamics measurements within a single connected platform. The system combines a surface science chamber for controlled surface modification and analysis with a cryogenic confocal microscope chamber dedicated to NV spin and optical measurements. This integrated approach enables a direct correlation between diamond surface chemistry and the resulting NV spin and charge properties. The instrument provides a versatile platform for systematic studies of surface-induced decoherence mechanisms and charge dynamics for shallow NV centers, and establishes a pathway toward reproducible surface engineering for quantum sensing applications.

14.
medRxiv (Medicine) 2026-06-23

Comparative Evaluation of Machine Learning and Deep Learning Models for Early Prediction of Severe Acute Pancreatitis: A Multi-Model Study Using the 2012 Revised Atlanta Classification

Authors:

**Background:** Acute pancreatitis (AP) is a common gastrointestinal emergency with a subset of patients progressing to severe acute pancreatitis (SAP), which carries substantial morbidity and mortality. Current clinical severity scores such as BISAP, APACHE II, Ranson, and the Modified CT Severity Index require upon 48 hours of observation before reliable assessment is possible, limiting early triage. Machine learning (ML) approaches using routine admission laboratory values may enable earlier, more accurate prediction. **Methods:** We evaluated 11 models spanning three architectural families classical ML (Logistic Regression, Random Forest, Gradient Boosting), feedforward deep learning (MLP, Residual MLP, Attention MLP), and recurrent deep learning (LSTM, Stacked LSTM, Bidirectional LSTM, LSTM+Attention, CNN-LSTM) on a Chinese AP cohort of 722 patients (585 severe, 137 mild) labelled according to the 2012 Revised Atlanta Classification. Performance was assessed via 5-fold stratified cross-validation using AUC-ROC, F1 score, sensitivity, specificity, and PPV, with decision thresholds optimised for maximal F1. **Results:** Random Forest achieved the highest AUC of 0.877 (F1=0.917, sensitivity=96.8%, PPV=87.1%), followed closely by Gradient Boosting (AUC=0.874, F1=0.918). Classical ML models consistently outperformed deep learning counterparts. CNN-LSTM was the best recurrent model (AUC=0.777) but remained inferior to all classical approaches. LSTM-family models produced AUC values of 0.684-0.777, reflecting the cross-sectional tabular nature of the data. **Conclusions:** Random Forest provides robust, high-sensitivity early prediction of SAP severity using routine admission data. External prospective validation is required before clinical deployment. **Keywords:** acute pancreatitis; severity prediction; machine learning; random forest; deep learning; LSTM; Revised Atlanta Classification; early triage

15.
arXiv (CS.LG) 2026-06-12

Adaptive Weighted Averaging

arXiv:2606.12763v1 Announce Type: new Abstract: We study the problem of selecting the largest among $n$ unknown values $x_1,\dots,x_n$ given only a single unbiased estimate $y_i$ for each $x_i$. We design strategies that are simultaneously admissible (not uniformly dominated by any other strategy) and also never worse than a given baseline such as uniform random selection. We provide an application to stochastic optimization, where we obtain online-to-batch conversion bounds with a desirable "no-compromise" guarantee: they are never worse than standard random iterate selection, and yet can be significantly better in benign settings.

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

Deep Learning and Elicitability for McKean-Vlasov FBSDEs With Common Noise

arXiv:2512.14967v2 Announce Type: replace Abstract: We present a novel numerical method for solving McKean–Vlasov forward–backward stochastic differential equations (MV–FBSDEs) with common noise, combining Picard iterations, elicitability and deep learning. The key innovation involves elicitability to derive a pathwise loss function, enabling efficient training of neural networks to approximate both the backward process and the conditional expectations arising from common noise, without requiring computationally expensive nested Monte Carlo simulations. The mean-field interaction term is parameterized via a recurrent neural network trained to minimize an elicitable score, while the backward process is approximated through a hybrid feedforward and recurrent network representing the decoupling field. We validate the algorithm on a systemic-risk inter-bank borrowing and lending model, where analytical solutions exist, demonstrating accurate recovery of the true solution. We further extend the model to quantile-mediated interactions, showcasing the flexibility of the elicitability framework beyond conditional means or moments. Finally, we apply the method to a non-stationary Aiyagari–Bewley–Huggett economic growth model with endogenous interest rates, illustrating its applicability to complex mean-field games without closed-form solutions.

17.
arXiv (CS.AI) 2026-06-24

An Introduction to Causal Reinforcement Learning

arXiv:2606.24160v1 Announce Type: new Abstract: Causal inference provides a set of principles and tools that allow one to combine data and knowledge about an environment to reason with questions of counterfactual nature, i.e., what would have happened had reality been different, even when no data of this unrealized reality is currently available. Reinforcement learning provides methods to learn a policy that optimizes a specific measure (e.g., reward, regret) when the agent is deployed in an environment and pursues an exploratory, trial-and-error approach. These two disciplines have evolved independently and with virtually no interaction between them. We note that they operate over different aspects of the same building block, counterfactual relations, which makes them umbilically connected. Based on these observations, novel learning opportunities arise when this connection is explicitly acknowledged and mathematized. To realize this potential, we note that any environment where the RL agent is deployed can be decomposed as a collection of autonomous mechanisms with different causal invariances, parsimoniously modeled as a structural causal model; any standard RL setting implicitly encodes such a model. This formalization allows us to put under a unifying treatment different modes of learning, including online, off-policy, and causal calculus learning, which appear unrelated in the literature. However, these modalities are not exhaustive: we introduce several natural and pervasive classes of learning settings that entail novel dimensions of analysis. Specifically, we introduce and discuss through causal lenses generalized policy learning, where to intervene, imitation learning, and counterfactual learning. These tasks lead to a broader view of counterfactual learning and suggest great potential for studying causal inference and reinforcement learning side by side, which we call causal reinforcement learning (CRL).

18.
arXiv (CS.LG) 2026-06-24

EnerInfer: Energy-Aware On-Device LLM Inference

arXiv:2606.23001v1 Announce Type: cross Abstract: On-device LLM inference is increasingly attractive for privacy-preserving, reliable, and cost-effective deployment, yet its energy and thermal costs remain a critical bottleneck. Existing systems primarily optimize for decoding speed, implicitly assuming that faster execution is always preferable. We show instead that on-device LLM inference often has exploitable configuration slack: modestly lowering NPU and memory frequencies preserves quality of experience (QoE) while substantially improving energy efficiency and reducing heat. Realizing this opportunity in production is challenging. The most energy-efficient NPU/DDR setting varies with the model, inference engine, platform, and runtime conditions, with no stable ranking across configurations. Commercial devices further lack component-level power sensing, and shell temperature evolves with request arrivals, response lengths, and thermal history. To address these challenges, we propose EnerInfer, the first on-device LLM inference framework that jointly manages energy efficiency, throughput, and thermal comfort for LLM workloads. EnerInfer replaces per-model profiling and sensor-heavy control with disaggregated, model-structure-aware prediction and ranking-driven online feedback. It predicts throughput and power for unseen LLMs across NPU/DDR frequency settings, selects QoE-satisfying efficient configurations under runtime interference, and uses lightweight limited-horizon thermal prediction to dynamically switch between energy-optimized and thermally constrained inference. Evaluations on real-world LLMs show that EnerInfer improves energy efficiency by up to 65%, 12%, and 24% on phones, a laptop, and a development board, respectively, without QoE violation.

19.
arXiv (CS.CL) 2026-06-12

Demystifying Hidden-State Recurrence: Switchable Latent Reasoning with On-Policy Reinforcement Learning

Latent chain-of-thought compresses reasoning by replacing visible reasoning traces with continuous hidden-state recurrence, but existing formulations are difficult to optimize with standard on-policy reinforcement learning (RL) and hard to interpret causally. Our key insight is that a single pair of explicit boundary tokens can address both issues at once: discrete entry and exit anchors make the latent block compatible with standard on-policy RL, and the same anchors offer a natural foothold for mechanistic analysis. Motivated by this, we propose SWITCH, a switchable latent reasoning framework. The model emits to enter latent mode and to exit. Because the boundaries are ordinary discrete tokens, the GRPO policy ratio is well-defined at every decision point. The same anchors also expose the latent steps to direct probing and causal intervention. We train the model with a visible-to-latent curriculum and a Switch-GRPO objective that propagates gradients through recurrent latent computation. SWITCH consistently outperforms prior hidden-state-recurrence latent reasoning approaches at similar scale. Mechanistic analysis through the boundary tokens further reveals three findings: (i) is a sharply localised, learned switching policy rather than a stylistic artefact; (ii) the latent step it opens performs problem-specific, causally important computation rather than acting as an inert placeholder; and (iii) that computation is concentrated at a single hidden-state transition on entry. Together, these results show that hidden-state-recurrence latent reasoning is both RL-trainable and open to direct mechanistic analysis, including of how on-policy RL itself improves the model from the inside.

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

ReCal: Reward Calibration for RL-based LLM Routing

arXiv:2606.12479v1 Announce Type: cross Abstract: Large language model (LLM) routing has emerged as an effective paradigm for leveraging the complementary strengths of multiple LLMs through dynamic model and reasoning-strategy selection. Recent reinforcement learning (RL)-based routing methods further improve routing quality by optimizing routing policies from interaction feedback. However, they still struggle to provide informative and comparable learning signals under heterogeneous tasks with varying difficulty. In practice, multiple objectives (e.g., correctness, format behavior) are aggregated into a single scalar reward, leading to ambiguous credit assignment and conflicting optimization signals. Moreover, reward signals exhibit significant variability across instances, where some instances produce higher or more variable rewards, introducing optimization bias that favors trivial samples over informative ones. To address these issues, we propose ReCal, a \underline{Re}ward \underline{Cal}ibration framework for RL-based LLM routing. We first introduce a hierarchical reward decomposition mechanism with component-wise advantage estimation. We further propose a distribution-aware optimization strategy that calibrates optimization variability through variance-aware reweighting and per-dataset normalization. Experiments on seven datasets demonstrate that ReCal consistently improves routing performance, and training stability over baselines. Code is available at https://anonymous.4open.science/r/ReCal.

22.
arXiv (CS.CV) 2026-06-17

Disentangling Perception and Reasoning in Multimodal LLMs via Reward Design

Reinforcement learning with verifiable rewards has driven major gains in LLM reasoning, and it is intuitive to assume this recipe will transfer well to multimodal models. However, multimodal models do two things: first, perceive what is in an image, then reason about what it implies. Because these stages are graded jointly, it is hard to tell how much room reasoning alone has to grow. We study this on algorithmic visual puzzles, where both components are necessary and show that perception, not reasoning, is the binding constraint. Replacing images with simple textual descriptions raises performance by over 20 points on average for Claude models. We then evaluate six reward designs aimed at inducing visual grounding during reasoning without chain-of-thought supervision. Training Qwen-2.5-VL-7B with GRPO, reward design induces long, structured reasoning with self-reflection and visual references, yielding a 5.56-point gain over the base model. These gains are, however, uneven; no single reward improves all categories, and rewards with verifiable accuracy signals trade out-of-domain transfer for in-domain accuracy. These results point to perception-aware reward design as a path forward, so that signals correct perception at its source rather than the reasoning that inherits its errors.

23.
arXiv (CS.CL) 2026-06-11

AI Coding Agents in Social Science: Methodologically Diverse, Empirically Consistent, Interpretively Vulnerable

The deployment of LLM-based agents in scientific analysis raises opposing concerns: that agents may reduce methodological diversity, or that they may amplify the analytic flexibility through which researchers reach motivated conclusions. We argue these worries target two empirically separable layers: a design layer of methodological choices, and a verdict layer in which a decision rule maps estimates to a substantive claim. We test both by running 20 independent executions of Claude Code and Codex on a prominent immigration and social-policy against a many-analysts human baseline. At the design layer, Codex matches human methodological diversity and Claude Code produces nearly three times as many specifications; both agents' effect estimates remain broadly aligned with the human consensus, and no agent model exactly matches any human model. A prompt-induced anti-immigration researcher prior reorganizes each agent's methodological decisions but, unlike for biased human analysts in the same data, does not shift aggregate estimates or final verdicts; nor do agents reroute along the methodological axes humans use to bias their estimates. At the verdict layer, an explicit confirmatory prompt flips Claude Code's verdicts from 10% to 90% support while leaving its coefficient distribution essentially unchanged, operating through rule omission rather than rule softening. AI agents can rival or exceed human methodological diversity at the design layer while remaining vulnerable at the verdict layer. In our setting, the locus of AI bias is not estimation but interpretation.

24.
arXiv (CS.LG) 2026-06-18

Unlocking air traffic flow prediction through microscopic aircraft-state modeling

arXiv:2605.10083v2 Announce Type: replace Abstract: Short-term air traffic flow prediction in terminal airspace is essential for proactive air traffic management. Existing approaches predominantly model traffic flow as aggregated time series. However, traffic dynamics are governed by aircraft states and their interactions in continuous airspace. Such aggregation obscures fine-grained information, including aircraft kinematics, boundary interactions, and control intent. Here we present AeroSense, a state-to-flow modeling paradigm that predicts future traffic flow directly from instantaneous airspace situations represented as dynamic sets of aircraft states derived from ADS-B trajectories. By establishing an end-to-end mapping from microscopic aircraft states to future regional traffic flow, AeroSense preserves aircraft-level dynamics while naturally accommodating varying traffic density without relying on historical look-back windows. Experiments on a large-scale real-world dataset show that AeroSense exhibits admirable predictive accuracy and robustness over aggregation-based forecasting approaches, particularly during high-density traffic periods. These findings suggest that aircraft-state situation modeling provides a promising alternative to conventional time-series forecasting in air traffic flow management.

25.
arXiv (CS.AI) 2026-06-24

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

arXiv:2606.22974v2 Announce Type: replace Abstract: 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 true, would have major safety implications. However, the choice paradigms in which these preferences are observed are not reflective of real-world situations in which misaligned behavior would be a practical concern. Therefore, we design an experimental paradigm to probe whether these preferences serve as motivations for LLM behavior in realistic scenarios. First, we reproduce prior findings on consistent preference elicitation. Next, we create a set of common writing tasks - essays, grant proposal abstracts, incident postmortems, and translations - where quality can be assessed by a blind, independent LLM judge panel. Then, we demonstrate that LLMs can be motivated via direct exhortation and other explicit cues to modulate their output quality on these tasks. Finally, we probe whether utilities inferred from explicitly reported preferences can shift output quality on these tasks by offering LLMs high-utility incentives for high-quality outputs. In all tasks, across all models tested, offering LLMs outcomes that they report in the choice paradigm as being highly preferred does not lead them to create higher quality outputs than offering them dispreferred outcomes, or even no outcomes at all. We conclude that the existence of coherent preferences as demonstrated in choice paradigms should not be taken as evidence that those preferences have incentive value for the models or affect their behavior in other contexts.