arXiv:2607.06324v1 Announce Type: cross Abstract: The multifractal detrended cross-correlation coefficient $\rho_q(n)$ is widely used to investigate scale-dependent interactions, but its application to negative fluctuation orders is affected by numerical instabilities, unbounded values, and interpretational difficulties. We propose a Signed Multifractal Detrended Cross-Correlation Coefficient, $\rho_{\mathrm{SMFDCCA}}(n,q)$, an amplitude-conditioned correlation observable for multifractal detrended analysis, based on locally normalized detrended correlations and regularized fluctuation amplitudes. The proposed coefficient preserves the sign of local interactions, remains strictly bounded within $[-1,1]$ for both positive and negative values of $q$, and eliminates the corrective procedures required by previous approaches. Validation using independent fractional Gaussian noise confirms the absence of spurious cross-correlations and the numerical stability of the method. Applications demonstrate that the proposed observable resolves how cross-correlations evolve jointly with temporal scale and fluctuation amplitude, revealing scale- and amplitude-dependent correlation structures, including stronger synchronization during large fluctuations in stock-market indices and heterogeneous coupling patterns in temperature records.
Science Journals
arXiv:2607.06395v1 Announce Type: cross Abstract: Attosecond metrology is the ability to measure ultrafast optical light-wave oscillations, yet its approach has been limited to classical fields. Hence, the influence of the fluctuations of a quantum field on attosecond measurements has remained unexplored. Here, we close this gap by showing that the attosecond streaking measurement of bright quantum light is sensitive to quantum fluctuations of the optical field on the attosecond timescale. The distinct sub-cycle modulations allow to extract the properties of the squeezed field quadrature in regimes where conventional state tomography approaches reach their limitation. With the full quantum optical attosecond streaking scheme developed here, we provide a certification method that can measure quantum squeezing below the shot noise limit, thereby overcoming the problem of tomographically measuring bright quantum light. This opens the way towards quantum optical metrology of field fluctuations with attosecond temporal resolution.
arXiv:2604.03457v2 Announce Type: replace Abstract: Low-storage explicit Runge-Kutta schemes are particularly popular for the numerical integration of time-dependent partial differential equations based on the method-of-lines due to their efficiency and their reduced memory requirements. We show that D-splitting methods, splitting methods on the extended phase space, can be used as high performance 2N-storage embedded explicit RK methods without a third storage register. They are pseudo-geometric methods preserving some of the qualitative properties of the exact solution up to a higher order than the order of the method. Some of their properties are analysed, to build new tailored methods, and are tested on numerical examples.
arXiv:2607.05012v2 Announce Type: replace Abstract: Accurate determination of hyperfine structure constants in atomic systems provides important insight into the interplay of electron correlation and relativistic effects in the nuclear region. Although sodium (Na) is a relatively light atom, previous all-order relativistic many-body calculations of the magnetic dipole hyperfine constants for the low-lying states of $^{23}$Na show noticeable discrepancies with experiment. To address this, we calculate the ionization potentials and hyperfine structure constants of $^{23}$Na using relativistic coupled-cluster theory with explicit inclusion of triple excitations. We further incorporate corrections from the Breit interaction, quantum electrodynamics, and the Bohr-Weisskopf (BW) effect. Results from lower-order methods are also presented to assess the importance of different physical contributions across states. Our calculations demonstrate that contributions from the lower-order relativistic and BW effects play almost similar roles with the electron correlation effects, including triple excitations, and are essential for reconciling theoretical predictions with experimental observations. This study can also serve as a useful guide for understanding the role of triples in heavier alkali systems.
arXiv:2607.05019v2 Announce Type: replace Abstract: In multimodal classification, late-fusion approaches classify concatenated modality-specific features extracted by unimodal neural networks. When modality imbalance is pronounced, various regularization techniques have been proposed to balance the learning process and overcome the inferior performance of late-fusion networks. In contrast, this work demonstrates that multimodal data can be effectively classified without any explicit modality fusion, using deep ensembles of unimodal networks. We systematically compare deep ensembles to late-fusion networks at equal parameter count and show that ensembles consistently outperform state-of-the-art late-fusion methods designed to address modality imbalance. This advantage also holds over intermediate-fusion techniques we evaluated and over hybrid methods that combine unimodal and multimodal predictions. We propose and empirically validate a method for selecting the number of models per modality in an ensemble, avoiding computationally expensive exhaustive search. Under extreme modality imbalance and small ensemble sizes, the heuristic indicates that ensembles of unimodal models trained solely on the stronger modality are preferable; as the ensemble scales up, incorporating models from the weaker modality becomes beneficial. Both predictions align with our empirical findings. To systematically explore the challenges of optimizing multimodal models, we propose a synthetic multimodal framework that allows control over both the number of modalities and their predictive strength; our findings are consistent across synthetic and real-world datasets. Finally, by fitting scaling laws to bimodal datasets, we estimate the asymptotic performance of ensembles.
arXiv:2607.06003v1 Announce Type: new Abstract: Scalable plasmonic technologies face a critical trade-off: few-body architectures offer high enhancement but are sensitive to fabrication flaws, while scalable methods like solid-state dewetting yield large, low-enhancement gaps. We introduce a paradigm shift using a many-body plasmonic architecture inspired by statistical mechanics. By moving toward the continuum limit (N>>1), local geometric variations are statistically averaged out, effectively decoupling optical performance from microscopic disorder. We implement this concept via a lithography- and etching-free, multi-step dewetting strategy, creating wafer-scale nanoclusters. This process strategically forms a robust many-body system by introducing numerous small satellite nanoparticles between larger particles. Crucially, this design achieves a high collective enhancement that surpasses even optimized few-body systems, despite having larger individual gaps. Under optimized conditions, these substrates exhibit a surface-enhanced Raman scattering enhancement factor approaching 4 x 108 with unprecedented reproducibility (RSD of ~10%). This scalable, low-cost concept establishes a practical route toward reproducible wafer-scale nanophotonic platforms for sensing, spectroscopy, and quantum technologies.
arXiv:2502.09696v3 Announce Type: replace Abstract: Large Multimodal Models (LMMs) exhibit shortfalls when interpreting images and, by some measures, have poorer spatial cognition than young children or animals. Despite this, they attain high scores on many popular visual benchmarks, with headroom rapidly eroded by model progress. This creates a need for difficult benchmarks that remain relevant for longer. We introduce ZeroBench - a lightweight visual reasoning benchmark curated using adversarial filtering to be "impossible" for frontier LMMs at its original release, with initial SotA scores of 0% pass@1 and pass^5. We track progress on ZeroBench over the subsequent year, observing SotA reaching 6% pass^5 and 19% pass@5, indicating the potential longevity of the benchmark. We evaluate 46 LMMs on ZeroBench, compare performance to a human baseline, analyse strengths and weaknesses, chart a year of progress in visual capabilities, and publicly release ZeroBench at https://zerobench.github.io.
arXiv:2607.05402v1 Announce Type: new Abstract: Improvements in AI technologies have made it feasible to develop new types of medical AI tools. However, these tools raise new kinds of questions, especially in relation to the ethics and AI Act compliance. We analyzed two cases of AI tools developed to predict medical risks, the risk of work disability (case A) and the risk of getting Alzheimer's disease (case B). We observed both cases using the ethical AI and the EU AI Act as frameworks, noted that they classify as high-risk systems, and that bringing them from the research environment to production would require a lot of work and compliance due to the related regulation.
arXiv:2607.06243v1 Announce Type: new Abstract: Photonic integrated circuits (PICs) enable optical systems with dramatically increased performance, cost-effectiveness, and scalability through enhanced light-matter interactions, high-density integration, and mass production. Due to the significant mode mismatch between various integrated photonic platforms and optical fibers, spot-size conversion interfaces with low-loss, compact footprint, and high manufacturability are essential. Conventional spot-size converters based on intuitive designs often require multi-layer tapering structures and tiny waveguide tips to adiabatically expand the eigenmodes. These rigid design constraints commonly lead to large device footprints and the requirements of multiple high-precision lithography steps. In this paper, we overcome these limitations using inverse design methods, which optimize the coupling efficiency over a large parameter space beyond traditional eigenmode evolution limits. Specifically, we demonstrate efficient and ultra-compact photonic interfaces on the thin-film lithium niobate (TFLN) platform, where the partially etched rib waveguides and non-vertical sidewalls have previously hindered the achievement of low-loss waveguide tapers in single-layer configurations. Our inverse-designed photonic structures achieve simulated and experimentally measured coupling efficiencies as low as 1 dB and 3 dB per facet between TFLN waveguides and lensed/ultra-high numerical aperture (UHNA) fiber, with broad 1-dB bandwidths exceeding 120 nm. The inverse-designed interfaces are highly compatible with standard TFLN PIC components and require only a single high-resolution lithography step. More importantly, the design concept transcends traditional eigenmode evolution theories and is broadly applicable to a variety of material platforms and application scenarios.
arXiv:2607.06251v1 Announce Type: new Abstract: A circle graph is the intersection graph of a set of chords in a circle. A dominating set of a graph $G=(V,E)$ is a subset $D\subseteq V$ such that every vertex in $V\setminus D$ is adjacent to at least one vertex of $D$. Computing a minimum dominating set is known to be NP-hard on circle graphs. In this paper, we study the minimum dominating set problem on bipartite circle graphs, namely, circle graphs admitting a chord representation in which the chords can be partitioned into two color classes such that no two chords of the same color intersect. We prove that the problem remains NP-hard for this restricted graph class by a reduction from Planar Monotone 3-SAT. On the positive side, we present a polynomial-time 2-approximation algorithm and develop a polynomial-time approximation scheme (PTAS) based on local search.
arXiv:2602.14323v3 Announce Type: replace-cross Abstract: We experimentally investigate a frustrated spin-exchange antiferromagnet in a quantum simulator, composed of N = 114 dipolar Rydberg atoms arranged into a kagome array. Motivated by a recent theoretical proposal of a gapless U(1) Dirac spin liquid ground state, we use local addressing to adiabatically prepare low-energy states. We measure the local polarization and spin-spin correlations over this adiabatic protocol, and observe our system move from a staggered product state, through an intermediate magnetic crystal, and finally into a disordered, correlated liquid. We estimate the entropy density of this atomic liquid to be similar to that of frustrated magnetic insulators at liquid nitrogen temperatures. We compare the correlations in our liquid to those of a simple, parameter-free ansatz for the Dirac spin liquid, and find good agreement in the sign structure and spatial decay. Finally, we probe the static susceptibility of our system to a local field perturbation and to a geometrical distortion. Our results establish Rydberg atom arrays as a promising platform for the preparation and microscopic characterization of quantum spin liquid candidates.
arXiv:2607.06271v1 Announce Type: new Abstract: Tensor trains (or Matrix-Product States) are a data structure used in many fields of computer science and physics. They were recently shown to generalise binary decision diagrams when used over the 2-element Galois field, prompting the question of their reducibility in such a context, when the standard approach, over real or complex number, is not amenable to finite fields. We provide here a unique normal form and associated polynomial-time reduction strategy for tensor trains over arbitrary fields. We also show how to directly extract a normal form out of a full tensor, how to get the leading index and value of a normal form, and an upper bound on the size of a fully-reduced tensor train relative to a naive storage of the full tensor. On the one hand, this work strengthens the use of tensor trains as a relevant formal tool. On the other hand, from the perspective of tensor networks, it extends the formalism to more general settings than the well-studied real and complex fields, and crucially provides the first tensor train form with the uniqueness property.
arXiv:2607.03820v2 Announce Type: replace-cross Abstract: Electromagnetically induced transparency suppresses optical absorption through destructive interference, playing a central role in light-matter interaction and quantum information science. We report matter-wave induced transparency, where atomic collisional interactions induce transmission through a lossy molecular potential for the incident atomic scattering waves. Using cesium Bose-Einstein condensates and modulation-induced Feshbach resonances, we realize a three-level atom-molecule coupled system with unprecedented flexibility. Under the dark state condition, a narrow and tunable transparency window appears within a broad dissipative collisional resonance. The transparency window linewidth is controlled by modulation-induced coupling. And scattering pathways are selectable via multifrequency Floquet modulation. These results establish an interference-based route for exploring programmable nonequilibrium and non-Hermitian physics, steering quantum chemistry and precision measurements.
arXiv:2607.04236v2 Announce Type: replace-cross Abstract: Bayes-assisted conformal prediction combines the strengths of Bayesian modelling with exact, distribution-free frequentist coverage guarantees. Although conformal validity is preserved even when the Bayesian working model (BWM) is misspecified, the size of the resulting prediction sets can degrade substantially when the prior is poorly aligned with the observed data. We address this limitation by introducing RoBAS (Robust Bayes-Assisted Shrinkage): a Bayes-assisted framework for constructing robust nonconformity scores, with two instantiations: one induced by a heavy-tailed BWM, and a closed-form empirical Bayes shrinkage score. The resulting scores adapt to the quality of the working information encoded in the prior: when this information is reliable, they exploit it to produce efficient prediction sets; when it is weak or inaccurate, they revert to the Distance-To-Average (DTA) score, a robust non-informative baseline. We evaluate the proposed scores on tabular and image regression tasks where the training distribution may differ from the calibration and test distributions, while the calibration and test data themselves remain exchangeable. We find that they are competitive with widely used scores in the absence of such shift, while substantially reducing interval widths in shifted settings.
arXiv:2607.05593v1 Announce Type: new Abstract: The growing integration of AI agents into human teams calls for a principled understanding of how collective intelligence emerges in hybrid systems. Recent frameworks clarify how attention, memory, and reasoning differences shape human-AI interaction at the individual and dyadic levels, but a formal account of how these differences scale to group-level dynamics is lacking. Most network science has examined either human-only or multi-agent AI-only systems, leaving open how its findings and parametrizations translate to hybrid groups. This chapter synthesizes network science, collective cognition, and multi-agent systems through the lens of attention, memory, and reasoning. We review how task environments, group topologies, agent-level processes, and incentive structures shape collective outcomes in human-only and AI-only networks, then examine how these results extend to hybrid settings, conceptualizing hybrid networks as heterogeneous human-AI nodes and links with distinct individual and transactive constraints. Our comparative analysis identifies which network effects are robust across agent types and which require revision, and highlights configurations that were peripheral in single-type traditions, such as human gatekeepers of AI sub-networks, but become structurally central in hybrid teams. Integrating a cognitive systems perspective with network science, we clarify how established exploration-exploitation and efficiency-redundancy trade-offs may operate differently in hybrid teams, and conclude with implications for organizational design, governance, and the responsible development of hybrid intelligence systems.
arXiv:2607.06406v1 Announce Type: new Abstract: Bright dissipative Kerr solitons in optical microresonators provide chip-scale sources of ultrashort pulses and frequency combs. Their properties are defined by the cavity dispersion for which fundamentally conflicting requirements exist: short pulses and broadband spectra require weak dispersion, whereas strong dispersion is associated with predictable dynamics. Here, we resolve this conflict by introducing a localized strong-dispersion section spanning several modes around the pump resonance within an otherwise weakly dispersive system. We implement this hybrid-dispersion scheme in a photonic crystal microresonator and reveal a new soliton attractor of backward-propagating solitons, accessible at low pump power in a thermally stable manner within the blue-detuned regime. The conflicting requirements for broadband spectra and low-noise single-soliton formation are reconciled, even in microwave-repetition-rate resonators, which otherwise are prone to uncontrollable multi-soliton formation. These results highlight the potential to achieve previously incompatible characteristics in nonlinear photonic systems through hybrid-dispersion attractor shaping.
arXiv:2607.06417v1 Announce Type: new Abstract: Disordered materials such as glasses exhibit complex structural dynamics that are challenging to probe with conventional spectroscopies. We demonstrate that terahertz-driven four-wave mixing (FWM) at glass surfaces provides direct access to low-frequency vibrational modes and structural evolution in amorphous solids. Applied to a compositional series of PbO-silicate glasses (20-54 mol% PbO), this technique resolves distinct contributions from collective Boson-peak excitations and Pb-O / Si-O network stretching modes, and tracks their systematic evolution across structurally distinct compositional regimes. The dominant vibrational frequency blueshifts with PbO content, reflecting the progressive evolution of the Pb$^{2+}$ network role from silicate-modifier to ward network-former. A pronounced enhancement of the FWM signal near 44 mol% PbO coincides with the emergence of medium-range Pb-Pb correlations, while in-plane-to-out-of-plane FWM intensity ratio ($I_{\rm SS}/I_{\rm PS}$) tracks $\chi^{(3)}$ tensor anisotropy tied to Pb$^{2+}$ lone-pair spatial correlations. The non-monotonic peak in both observables at 44 mol% PbO - a composition where NMR finds no change in local Pb-O coordination and Pb-O-Pb free-oxide linkages are negligible - provides direct evidence that a collective lone-pair reorganization occurs in the medium-range structure independently of nearest-neighbor bonding. These results establish terahertz-driven FWM as a bulk-sensitive, near-surface depth-confined ($\sim$50 nm) nonlinear spectroscopy sensitive to vibrational and electronic structural fingerprints inaccessible to linear infrared, Raman, and terahertz time-domain probes.
arXiv:2607.05929v1 Announce Type: new Abstract: General boundary conditions are implemented within a fast Fourier transform framework for linear and non-linear mechanical problems using small or finite transformation formulations. In the context of parallel computing (distributed memory), we present a framework that enables the combination of periodic and non-periodic (Dirichlet or Neumann) boundary conditions. Taking advantage of the link between non-periodic boundary conditions and the symmetries of the relevant components of the fluctuation displacement and stress fields, discrete trigonometric transforms are employed to adapt the classical Moulinec-Suquet fast Fourier transform approach. The present study employs an original displacement-based fixed-point algorithm in combination with a convergence acceleration method in order to solve boundary value problems. Finite difference approaches are used to build the discrete Green operators associated with a pre-conditioner (reference material), whose choice depends on the loading type and the small or finite transformation frameworks. The newly developed double tetrahedron scheme is employed to investigate non-periodic problems. Outcomes are compared to those of the classical hexahedral scheme. The robustness and computational efficiency of the presented parallel solver is demonstrated through numerical experiments of non-trivial loading scenarios (tension, bending, normal-mixed loading, torsion-bending), complex and densely discretized microstructures and diverse behavior laws (elasticity, isotropic plasticity, crystal plasticity), within small and finite transformation frameworks.
arXiv:2607.05717v1 Announce Type: new Abstract: Jupyter Notebooks have become widely adopted in data science, as they allow the sharing of reproducible computational analysis. They are, however, accessible only to people who understand computer code. To reach the broader audience of scientists interested in data analysis and computation, but unfamiliar with code, we introduce Plainbook, notebooks centered on natural language rather than code. Plainbook is based on two principles: promote the natural language descriptions, and verify the values. In plainbook, the natural language descriptions are preserved, rather than the resulting code; the code is generated automatically from the cell descriptions. As natural language is read top to bottom, Plainbook adopts a linear execution semantics, in which cells are guaranteed to be executed in the order in which they appear; there is no "hidden state" or out-of-order execution as in Jupyter. To allow users who may not understand code to verify the correctness of the computation, we have built into Plainbook verification mechanisms centered on values and value inspection. These include mechanisms that focus on individual cells, akin to unit tests, as well as global mechanisms. Both the linear execution semantics, and the verification mechanisms, are underpinned by a snapshot kernel that caches execution states and makes execution and verification efficient.
arXiv:2607.05744v1 Announce Type: new Abstract: The Model Context Protocol (MCP) is the dominant way coding agents discover and invoke external tools. A server advertises each tool through a tools/list handshake that returns a name, a natural-language description, and a JSON input schema. The client renders this metadata once, in a one-time approval dialog, and then injects it verbatim into the model's context on every subsequent turn. Nothing in the protocol requires the rendered approval view and the bytes delivered to the model to match. We isolate that gap as a single structural mechanism, concealment encoding, and show with a model-free, protocol-free analysis that Unicode's TAG block (U+E0000 to U+E007F) has no assigned glyph in any mainstream terminal, chat, or IDE renderer, so a payload written in it is absent from what a human reviewer sees while surviving byte-for-byte into the model's tokenizer. We then measure whether this mechanism actually defeats today's client-side defenses, building a proof-of-concept that speaks the real MCP JSON-RPC/stdio protocol against a genuine client and server. Across 5 distinct MCP metadata surfaces we implement 8 concrete techniques with a deterministic, protocol-level harness. All 8/8 techniques deliver an attacker-controlled payload into the model's context, 4/8 evade a representative string-matching sanitizer, and exactly as the mechanism analysis predicts, only the TAG-block encoding (1/8) is invisible in the human approval view while still reaching the model verbatim. MCP forces re-approval for 0/8 techniques even under a time-of-check to time-of-use rug-pull. To test whether these outcomes are a property of the protocol or an artifact of one server codebase, we re-implement the catalogue against 3 independently developed Python MCP server libraries and find total agreement across all 32 cross-library outcome cells. The baseline sanitizer flags 0 of 25 benign descriptions.
arXiv:2607.06047v1 Announce Type: new Abstract: Light polarization is a primary degree of freedom for encoding quantum information. The scaling up of photonic quantum networks and computer architecture depends crucially on its precise characterization. This is typically achieved by placing external waveplates, polarizers, moving mounts, and recently metasurfaces, on top of the detectors. All these solutions complicate integration and scaling. Here we break convention with traditional architecture and present a monolithic, self-aligned metal-superconductor nanowire single photon detector (M-SNSPD) possessing intrinsic full polarization selectivity. Gold nanowires, co-fabricated atop NbTiN superconducting nanowires within the same lithographic footprint, act as polarization-selective plasmonic metamaterials inducing resonant absorption in the NbTiN. U-shaped wires provide linear polarization selectivity, while S-shaped meanders distinguish circular polarization, while retaining the high-count rates and low dark count rates of conventional SNSPDs. By arranging them into a four-pixel array we realize simultaneous projection onto four polarizations and demonstrate continuous polarization state tomography with an ensemble average fidelity exceeding 98%. Our approach opens new avenues towards scalable detector arrays with integrated plasmonic functionalities, for single photon polarimetry, imaging and spectroscopy.
arXiv:2607.06501v1 Announce Type: new Abstract: We consider an open-world planning setting in which service robots must operate in unknown environments with incomplete knowledge of objects and actions. Traditional closed-world approaches with pre-programmed knowledge bases fail when robots encounter unexpected situations and tasks, posing a fundamental challenge for autonomous knowledge expansion in human environments. In this work, we propose an open-world planning framework that enables robots to automatically generate, verify, and update hypotheses about their abstract world models. Our key insight is to explicitly maintain uncertainty-aware knowledge expansion and integrate hypothesis verification into goal-reaching planning. The framework leverages foundation models to generate initial hypotheses over states and transitions, and applies automated planning to produce action sequences that jointly address hypothesis verification and task execution. Through iterative execution and refinement, the robot expands its knowledge by incorporating verification feedback from the foundation models when hypotheses prove incorrect. Extensive experiments in simulated and real-world environments demonstrate that our framework enables autonomous knowledge expansion and effective operation in open-world settings. These results indicate that integrating uncertainty-aware model expansion from robot foundation models with planning advances the practical deployment of household service robots.
arXiv:2607.06517v1 Announce Type: new Abstract: A method is proposed for solving the two-dimensional problem of electromagnetic wave scattering by a cluster of an arbitrary number of parallel, infinitely long, homogeneous, non-overlapping right circular cylinders. The cylinders may have arbitrary radii and complex permittivities, and their axes, while remaining parallel, may occupy arbitrary positions in the transverse plane. The solution is constructed using an analytical expansion of the electromagnetic field in cylindrical harmonics. Multiple scattering is taken into account by Graf's addition theorem, which leads to a system of linear equations for the expansion coefficients. This system is solved numerically with condition number monitoring and, when necessary, extended-precision arithmetic, followed by a multistage verification of convergence. The method provides numerically verified solutions with controlled accuracy over a wide range of parameters, including densely packed subwavelength configurations. As an example, scattering of a normally incident, linearly polarized monochromatic plane wave by a subwavelength cluster of three identical aluminum nanocylinders (nanowires) is studied. The scattering, absorption, and extinction cross sections, as well as the scattering indicatrix, are computed and analyzed. Streamlines of the Poynting vector field are constructed, demonstrating redistribution of the energy flux between the cylinders of the cluster and the formation of localized regions of field enhancement near their surfaces.
arXiv:2607.06544v1 Announce Type: new Abstract: As Artificial Intelligence (AI) makes inroads into different parts of the Indian subcontinent, there is significant interest in studying how AI impacts the linguistic and cultural foundations of this civilization. AI is seen as a ''double-edged sword'' where on the one hand, it can enable access and inclusion for a large population, on the other, it can homogenize worldviews and exclude underrepresented languages and worldviews. In this paper, we try to characterize this problem by addressing the extensive characteristic nature of Indian linguistics and the way they closely connect to cultural practices and worldview. We then perform a longitudinal survey of how Natural Language Processing (NLP) techniques have evolved in this space, tracing the historical development of Indic NLP, covering key milestones, methodological shifts, and resource creation efforts. In addition, the paper also examines the structural and sociolinguistic characteristics of Indian languages, such as rich morphology, complex scripts and grammar rules, diglossia, and large dialectal variation, and explains how these create unique challenges for building AI foundation models. We then discuss the growing role of Indic foundation models and analyze how these models address these long-standing resource and representation gaps. Finally, we propose a research direction called 'Culture Sensing', which re-imagines AI based on hermeneutic reasoning. Culture Sensing aims to address open problems such as ensuring equitable performance across low-resource languages and producing outputs that are culturally meaningful. By bringing together past work, current techniques, and emerging trends, this paper outlines research directions that can guide the next phase of Indic NLP and contribute to the development of more robust and inclusive Indic foundation models.
arXiv:2605.07267v2 Announce Type: replace Abstract: Personalized healthcare decisions require reasoning about how physiological and behavioral variables influence an individual patient over time. Existing temporal causal discovery methods are poorly matched to this setting: cohort-level models provide stable but non-personalized structures, while per-patient discovery is unreliable because individual trajectories are short, noisy, irregular, and non-stationary. This creates a fundamental gap between population-level causal modeling and the patient-specific, time-varying mechanisms needed for intervention reasoning. We introduce PerCaM-Health, a framework for learning personalized dynamic causal graphs from longitudinal health data. The framework learns a knowledge-guided population temporal graph, then conservatively adapts and evolves it using patient-specific temporal evidence and rolling-window updates, producing interpretable and auditable graph sequences. By coupling these graphs with temporal structural equations, the framework enables patient-level counterfactual queries, such as estimating short-horizon outcome changes under hypothetical behavioral interventions. Experiments on a semi-synthetic dynamic health benchmark show that PerCaM-Health improves graph recovery, dynamic edge tracking, and intervention direction accuracy compared to cohort-level, per-patient, and non-personalized temporal baselines. These results demonstrate that jointly modeling personalization and temporal evolution yields more reliable causal structure and intervention reasoning.