Robotics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Friday 4 September
We study dark manipulation: after a brief lit Write encodes z0 = Enc(rgb), a policy pi(z) and open-loop dynamics f(z,a) complete contact-rich skills without further pixels (dark_f). On ManiSkill StackCube (n=160; seed packs 0/1000), dark_f attains 68.1% stacked on the five-rung chain (near_A -> grasped -> lifted -> on_B -> stacked), compared with 35.6% for per-step lit_reenc and 0% for freeze/encode_black. On a shared Write->HOLD protocol (n=40), occlusion and camera-aligned GT contact-neighbor masks drive lit lift from 43% to 0%, while dark_f holds 82.5%; shuffling actions inflates dynamics MSE by ~9.4x; write-time appearance shifts break encoding (night: 0% stacked), yet the same shifts during HOLD leave dark_f lift unchanged; Write length Tw is flat once the stop phase is reached, while earlier stops and write-time blur/JPEG sharply cut stacked. A dedicated pi_write reaches 35% vision-budget stacked (n=80); matched Dreamer-style/pixel nulls without privileged geom stay at 0%. Privileged state-RSSM MPC reaches ~35% stacked with 9D dark observations – a stronger-observation null, not a matched visual baseline.
Digital Libraries
Evidence Records for Public-Record Changes: Definitional Class Membership as the Certifiable Citation Unit
Public records change in ways that matter for later interpretation, yet citing a change precisely is difficult–most consequentially, separating what the change demonstrably is from what it is claimed to mean. Definitional class membership–whether a change matches an objectively checkable predicate–is certifiable: it can be independently verified against the cited source. Importance or review-priority did not certify in our review design: competent reviewers did not converge on it. We introduce the Evidence Record, a deterministic, hash-pinned, claim-bounded primitive that makes a public-record change citable through a record whose integrity any reader can verify, and that carries its definitional class memberships explicitly. Integrity is checkable from the published package alone; deciding most class predicates additionally requires registry state that records identify by hash but do not embed. We implement it in TrialDiff for ClinicalTrials.gov protocol amendments and report both directions of the boundary. On the certifiable side, deterministic event-classes over a fixed, non-random corpus of 100 breast-cancer-related interventional trials yield citable records; a naive "primary endpoint changed after completion" flag fires on 73 patches, but 63 co-occur with results posting and only 10 are clean–a separation made as a checkable definitional predicate, not a validated priority judgment. Two defects found after publication, in opposite directions, were corrected by reissue rather than rewriting. On the uncertifiable side, a derived severity/priority label could not be certified under blinded adversarial review (reviewer confirmations of 4 to 17 of 30 against a pre-registered 24/30 gate) and is released only as uncalibrated metadata. The certifiable citation unit for a public-record change is definitional class membership, not a priority ranking.
Read the full abstract
Artificial Intelligence
Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work
Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations. Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability is delivered. Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning. We present Occamy-1.0, a cost-efficient co-work model obtained by further training the post-trained Qwen3.6-35B-A3B checkpoint. We construct execution-grounded data and environments, capture replayable long-horizon trajectories across multiple harnesses, and use staged post-training to develop and consolidate complementary execution capabilities. Across a broad suite of co-work benchmarks, Occamy-1.0 is consistently among the strongest comparably sized models and remains competitive with substantially larger frontier systems on several tasks. Under our stated evaluation and pricing protocol, its aggregate performance across four representative benchmarks places it at the low-cost knee of the observed cost–performance Pareto frontier. Supporting evaluations in tool calling, coding, and instruction following further show that this specialization preserves broad agentic capability. We release the model weights and a subset of the training data to support research on practical co-work agents and agentic post-training.
Read the full abstract
General Physics
Einstein and the Solar Eclipses - Why Albert Einstein joined the Astronomische Gesellschaft in Potsdam 1921
In 1911 Einstein suggested that photons which pass close to the Sun are deflected by its effective mass which would apparently shift the positions of stars by up to one arcsecond, something that should be observable by astronomers during solar eclipses. Einstein`s ideas including the later doubling of the value due to relativistic space curvature met with enthusiastic responses from astronomers. The history of these efforts, their motivations and interactions with Einstein is presented here, beginning with the expedition that failed due to WWI of the young Erwin Freundlich in 1914 and ending with the unsuccessful Potsdam campaign to Öland (Sweden) in 1954. A newly found conference photograph from the first post-war meeting of the Astronomische Gesellschaft in 1921 shows Albert Einstein as a new member of the AG on the Telegraphenberg in Potsdam.
Read the full abstract
Game Theory
Strategic Information Transmission over Gossip Networks
We consider a fully connected gossip network of $n$ nodes that track a binary continuous-time Markov source through a strategic sender transmitting updates under a communication budget at a rate that depends on the source state. The receivers exchange packets through gossip and decide whether to follow the sender. We model this interaction as a Stackelberg game and analyze it through a stochastic hybrid systems (SHS) framework. We prove that the sender's budget constraint binds at every interior optimum, reducing its problem to a one-dimensional search on the budget line. When the sender pushes its preferred state at the higher rate, the receivers gossip at the highest available rate. Gossip has no direction of its own and works against the asymmetry in the sender's policy rather than reinforcing it. We prove that an optimistic Stackelberg equilibrium exists, and that it is unique and explicitly characterized whenever a policy on the strategic half of that line is feasible at the gossip cap. Monte Carlo simulations agree with the analytical recursion.
Read the full abstract
General Relativity and Quantum Cosmology
Initial conditions for inflation
We propose a single-field mechanism that dynamically prepares the initial conditions required for exponential plateau inflation. The inflaton is decomposed into a coarse-grained mode and a high-occupancy, approximately Gaussian band of relativistic nonzero modes. Gaussian–Hartree averaging of the potential produces an exact variance-dependent minimum, which places the coarse-grained field high on the plateau when the excitation variance is large. Under a translation-invariant microcanonical measure, the available phase-space volume is exponentially concentrated near this minimum, while an order-one homogeneous kinetic-energy fraction is exponentially atypical. During expansion, the preparation modes redshift as radiation, $\sigma^2\propto a^{-2}$ and $\rho_\chi\propto a^{-4}$. Although the induced minimum rapidly moves toward the vacuum, the force acting on the coarse-grained mode is bounded by an exponentially small factor, yielding analytic upper limits on its velocity, displacement, and kinetic fraction. We derive a sufficient condition for the subsequent number of slow-roll e-folds and exhibit a benchmark with more than 60 e-folds. The mechanism also motivates correlated finite-onset signatures: large-scale power suppression with decaying oscillations and a variance-convolution bispectrum.
Read the full abstract
High Energy Physics (Phenomenology)
The Masses of Bosons and Usual Fermions on Supersymmetric $SU(3)_{C}\times SU(4)_{L}\times U(1)_{X}$ Model
Rings and Algebras
Local derivations is a Lie algebra
In this paper, we utilize the reflexive hull to unify the concepts of local and almost inner derivations. By investigating the algebraic structure of reflexive hulls, we prove the conjecture on the structure of local derivations proposed by Ayupov, Elduque, and Kudaybergenov (J. Pure Appl. Algebra, 2023), as well as the conjecture concerning the structure of almost inner derivations proposed by Burde, Dekimpe, and Verbeke (J. Algebra Appl., 2018).
Read the full abstract
Nuclear Theory
Hadron resonance gas with density-dependent interactions for neutron stars and heavy-ion collisions
We present a density-dependent generalization of the van der Waals hadron resonance gas model (DD-HRG) for describing both the hot hadronic matter created in heavy-ion collisions and the cold, dense matter inside neutron stars. Non-resonant interactions are incorporated through a generalized excluded-volume prescription with a density-dependent available-volume fraction, supplemented by an arbitrary density-dependent mean field. With isospin-dependent interaction parameters constrained by empirical properties of nuclear matter, the resulting equation of state extends the causality range to include neutron-star interiors and supports two-solar-mass stars. It also improves the description of lattice QCD thermodynamics and conserved-charge susceptibilities at vanishing baryochemical potential, with lattice data favoring reduced repulsion among strange baryons. This DD-HRG framework is available within the latest version of the open-source Thermal-FIST package.
Read the full abstract
Artificial Intelligence
DI-Bench: Systematically Generating In-Domain Data Intelligence Benchmarks for Enterprise Agents
Evaluating enterprise agents on domain-specific benchmarks is critical, yet public benchmarks rarely evaluate whether agents can integrate business knowledge with analytical computation, and constructing such benchmarks manually is costly. We present DI-Bench, a pipeline for generating realistic benchmarks for data intelligence (DI), the practice of extracting insights from large volumes of enterprise data. To emulate realistic DI tasks that require both computation and knowledge retrieval, DI-Bench builds an artifact linkage graph over data tables, dimensions, metrics, and documents to form questions involving structured data and associated knowledge. Ground truth answers are derived via query execution, followed by LLM question generation and validation. Applied to two public datasets, the pipeline produces a 731-task benchmark covering knowledge retrieval, analytical computation, and rule-grounded reasoning. To show the discriminatory capability and difficulty of the benchmark, we evaluate four models, revealing a substantial finding: models achieve only 32% accuracy when doing computational tasks where retrieved business rules modify the computation.
Read the full abstract
Optimization and Control
Convexity from order: a theory of faces and generalized separation
It is known that any convex cone in a general vector space is the positive cone of a preorder compatible with the vector space, and that any convex set is an affine slice of a convex cone. We therefore treat convex sets order-theoretically, as affine slices of positive cones. We find that faces are sets of strong minima, and we generalize functionals to oriented maps valued in arbitrary ordered vector spaces. Oriented maps expose every face, and an oriented affine map $f$ separates $C$ from $D$ when $f(D) \le 0 \le f(C)$ element-wise. This generalization of hyperplane separation recovers mathematical objects that are "local", including the generated face at a point and the normal cone. The kernels of separating oriented maps form a lattice of separating affine flats whose bottom element is the joint supporting subspace (JSS), recently introduced in finite dimensions for qualification-free convex analysis results. We define the JSS in general vector spaces for two convex sets $C, D$ as the affine slice of the cross-lineality $\mathrm{lin}(K-G)$, where $K,G$ are the homogenization cones of $C, D$. We show that in finite dimensions, bilateral facial reduction corresponds to a descent in the lattice of separating affine flats, which we formalize as lexicographic products of oriented maps. The lexicographic characterization of faces follows as a corollary. By making no assumption other than convexity, our approach yields general results: we show that the face of a difference of two convex sets is a difference of faces, dropping assumptions of compactness, finite dimensionality, and exposedness. We also show that any face of an intersection of two convex sets is an intersection of faces, resolving open problem 5.10 by Weis in "A note on faces of convex sets", dropping the assumption of non-empty intrinsic core.
Read the full abstract
Artificial Intelligence
Inference-Time Graph Engineering for Multi-Agent LLM Workflows
Recent multi-agent LLM systems increasingly rely on graph-structured communication to coordinate specialized agents. We revisit multi-agent orchestration from a graph-engineering perspective: rather than optimizing a static topology, we synthesize a task-conditioned temporal workflow graph that jointly specifies agent connectivity and edge-level communication semantics. We introduce ReActNet, a training-free framework that compiles a query and a set of role-specialized agents into a sequence of directed communication graphs. Each graph snapshot corresponds to one reasoning stage, and each edge carries a natural-language instruction specifying the message that a source agent should provide to a target agent. The compiled temporal graph is then executed through structured message passing: agents update their reasoning states by integrating their previous states with messages from controller-assigned neighbors, and a final aggregator synthesizes the resulting states into the answer. This design separates graph compilation from graph execution, making multi-agent coordination explicit, inspectable, and task-conditioned without requiring reinforcement learning or gradient-based topology optimization. Across knowledge reasoning, mathematical problem solving, code generation, and GAIA-style assistant tasks, ReActNet consistently improves over fixed-topology and learned-topology baselines while maintaining competitive inference cost. These results suggest that effective multi-agent orchestration depends not only on which agents communicate, but also on engineering executable workflow graphs that encode when, why, and how information should flow during reasoning.
Read the full abstract
High Energy Physics (Phenomenology)
Fermion mass hierarchy and flavor signals from a $D_4$ dihedral symmetry within a 2HDM-Tx with four-zero textures
We propose an extended Texturized Two-Higgs Doublet Model (2HDM-Tx) where a discrete $D_4$ symmetry predicts a four-zero texture in the fermion mass matrices. In contrast to the standard ansatz where the underlying Yukawa matrices are assumed to be structurally identical, we introduce a novel configuration of non-equivalent matrices that complementarily generate a specific Hermitian four-zero mass structure. This framework naturally suppresses flavor-changing neutral currents and provides a structural rationale for the fermion mass hierarchy, while in addition including a dark matter candidate. As a concrete phenomenological application, we show that the model simultaneously accommodates the persistent neutral resonance excesses at $95\text{ GeV}$ and $152\text{ GeV}$, satisfies current \text{LHC} Higgs data, and is consistent with the stringent upper bounds from lepton flavor violation processes. These combined constraints severely restrict the parameter space of the scalar phenomenology. Finally, we study the prospects for the discovery of the lepton-flavor violating process $h \to \tau \mu$ at the High-Luminosity LHC.
Read the full abstract
Statistical Methodology
Flexible Transformations for Bayesian Score Calibration
Modern statistical models are growing increasingly complex in an effort to realistically capture system dynamics. Using standard simulation-based inference, these models may be computationally prohibitive, necessitating the use of model calibration methods. Bayesian score calibration is a computationally efficient framework for model calibration with strong theoretical guarantees. This framework learns an appropriate correction for an approximate model using a small number of simulations from the data-generating process. Currently, only a location-scale transformation has been explored, which may lack the flexibility to correct the complex error introduced by some approximate models. In this paper, we develop two flexible transformations for use in the Bayesian score calibration framework. The first is a polynomial extension, which can appropriately adjust approximate models with location-varying error. The second is a sequential application of Bayesian score calibration, which can accommodate approximate models with posteriors that have low support for the true parameter values. We also discuss an additional diagnostic for use with this framework. We demonstrate the increased flexibility these two approaches provide over Bayesian score calibration in two illustrative simulation studies.
Read the full abstract
Information Theory
Moments of crosscorrelation demerit factors of binary sequences
Families of sequences with low mutual aperiodic crosscorrelation assist the design of systems for multi-user asynchronous communications and multiple-input multiple-output radar. The crosscorrelation demerit factor of a pair of sequences is the sum of the squared magnitudes of their crosscorrelation values at every shift when the sequences are normalized to unit Euclidean norm, and the merit factor is the reciprocal of the demerit factor. For each positive integer $\ell$, we endow the $2^{2 \ell}$ pairs of binary sequences of length $\ell$ with uniform probability measure and study the distribution of their crosscorrelation demerit factors. Sarwate showed that the mean value is always $1$ regardless of length $\ell$. We develop a method for finding an exact formula for the $p$th central moment (for any positive integer $p$) as a function of $\ell$. Formulae for the variance and third central moment ($p=2$ and $3$) are then obtained by hand calculations, while the fourth through sixth central moments are obtained by computer-assisted calculations. Our theory also shows that all the central moments must be strictly positive for $p\geq 2$ and $\ell \geq 3$.
Read the full abstract
Machine Learning
RAPTOR: Role-Aware Private Training for Mixture-of-Experts
Differentially private (DP) fine-tuning methods treat sparse Mixture-of-Experts (MoE) models as a single dense block, ignoring that shared layers see all data while experts only see routed records. We identify and formally characterize three resulting failure modes: global clipping suppresses expert gradients, batch-level normalization dilutes sparse expert updates, and fixed privacy noise degrades signal-to-noise ratio on low-load experts. We introduce RAPTOR - a Role-Aware Private Training framework, which alternates shared and expert optimization and targets each failure directly, using expert-specific clipping and noise together with a public expected-owner denominator and a count-independent update schedule that avoids conditioning on private, realized expert counts. We prove the resulting mechanism satisfies $(\varepsilon,\delta)$-DP: because each record is assigned to exactly one owner expert, per-expert mechanisms within a layer compose in parallel, so updating all $E$ experts costs no more, in privacy terms, than updating one, with shared and expert streams composing sequentially across training. We further derive a bias-variance decomposition of the public-denominator estimator showing its bias grows predictably with routing imbalance, yielding a privacy-free rule for selecting which layer to protect from routing entropy measured on a small public corpus. Experiments on Switch Transformer and OLMoE fine-tuning across GLUE tasks, and on DeepSeek-VL2-Tiny, show consistent gains over standard DP baselines across several privacy levels ($\varepsilon$), with the largest margins typically at the tightest budgets. Code and models are publicly available: https://github.com/leduckhai/RAPTOR
Read the full abstract
Dynamical Systems
Uniform in time propagation of chaos for noisy mean-field coupled maps
We study discrete-time $N$-dimensional mean-field systems on the torus subject to additive noise whose probability density is bounded away from zero. Given a Lipschitz one-particle map and interaction term, we prove that, if the lower bound on the noise density is sufficiently large, the $N$-dimensional transfer operator $\mathcal P_N$ preserves a dimension-independent class of sub-Gaussian probability measures. Moreover, $\mathcal P_N$ is a one-step contraction in the Dobrushin-Wasserstein distance and therefore admits a unique invariant measure $\rho_N$, which is itself sub-Gaussian. Under the same condition on the noise strength, we show that the associated self-consistent transfer operator is a one-step contraction in Wasserstein distance and hence admits a unique fixed point $\rho$. We further prove, using these contraction estimates, that $\mathcal P_N$ preserves an $O(N^{-1/2})$ neighbourhood of the product measure $\rho^{\otimes N}$ in the Dobrushin-Wasserstein metric, and in particular that $\rho_N$ belongs to this neighbourhood. Finally, we establish uniform-in-time propagation of chaos in the Dobrushin-Wasserstein metric and, under the additional assumption that the noise density is of bounded variation, in total variation for every fixed-dimensional marginal.
Read the full abstract
Complex Variables
Holomorphic Maps from $\mathbb{C}^p$ into Semi-Abelian Varieties
We prove the Second Main Theorem with truncation level one for holomorphic maps with Zariski-dense image intersecting a reduced effective Cartier divisor: Let $A$ be a semi-abelian variety with an equivariant compactification $A\subset\overline A$, and let $f:\mathbb C^p\to A$ be a holomorphic map with Zariski-dense image. Let $D$ be a reduced effective divisor on $A$ extending to a divisor $\overline D$ on $\overline A$. After possibly replacing $\overline A$ by another equivariant compactification depending only on $D$ and independent of $f$, for every $\varepsilon>0$, \[T_f(r,\overline D)\leq_{\mathrm{exc}}N_f^{[1]}(r,D)+\varepsilon T_f(r,\overline D).\]
Read the full abstract
Human-Computer Interaction
Will My Assistant Remember My Allergy? What Personal LLM Assistants Forget When Conversation Memory Is Compressed
Personal LLM assistants (health companions, elder-care agents, accessibility aides) are judged by what they remember about a person: a medication or an allergy mentioned in passing and needed days later. Privacy pushes them on-device, where a month of conversation can outgrow the model's own weights, so an eviction policy must decide what the cache forgets. Benchmarks report that eviction keeps such facts at a 20% budget, but they compress a prompt that already contains the user's future question, foresight no cache-reusing assistant has. Hide the question until after compression and the advantage vanishes: on PA-Bench, 100 assistant conversations we construct, an allergy mentioned in passing survives to the question that needs it 0–1% of the time, against 97% with full memory. The cause is the budget, not the scorer: none of the training-free policies we evaluate ranks the fact high enough, and the budget that would keep it is too large to bother compressing. A compressed cache is an inference-reuse mechanism, not a persistence layer: safety-critical facts need an auditable episodic store alongside it, and an interface that asks rather than invents.
Read the full abstract
Machine Learning
Data Scout: Targeted Web Crawling for Domain-Specific Pretraining Corpora
The dominant approach to building domain-specific pretraining corpora is to filter large web archives such as CommonCrawl. This works well for popular domains but breaks down for specialized ones, where relevant content is sparse and often beyond the reach of popularity-driven crawlers. We present Data Scout, which inverts this: instead of filtering an archive, it directs a targeted crawl. An LLM expands a root topic into a taxonomy and thousands of search queries; the returned URLs (seeds) are grouped by subdomain and screened with a user-supplied classifier (the probe), admitting each subdomain on the basis of a small sample. This works because relevance has a sharp boundary at the subdomain level: in mathematics, a page is 21x more likely to be relevant than one on a sibling subdomain. With the FineMath classifier as the probe, 21.9% of crawled pages are high-quality math content, 70x the 0.31% rate from filtering a comparable web sample, so the crawl wastes far less effort. But the payoff is not just efficiency: 63.2% of these pages are missing from CommonCrawl altogether, yet just as useful for training. Continued pretraining of Llama-3.2-3B on 1.9B Data Scout tokens matches FineMath corpus on GSM8k. Because the probe is the only domain-specific component, Data Scout can in principle apply to any domain with such a classifier.
Read the full abstract
Game Theory
Utility-Driven Spatial Data Sampling for UAV-Assisted Scientific Smart Farming
Large smart-farming deployments generate continuous scientific data from spatially distributed sensors, including soil, humidity, temperature, crop-health, and pest-related measurements. In vast agricultural fields, however, an energy-constrained unmanned aerial vehicle (UAV) often cannot collect data from every sensor during each mission. Existing UAV-assisted collection methods typically optimize coverage, route length, data volume, or freshness, but they do not always distinguish between data that is merely available and data that is scientifically valuable. This poster introduces a utility-driven spatial sampling framework for UAV-assisted smart farming. The field is partitioned into grid cells, each sized according to the UAV ground coverage range. After an initial exploration phase, each cell receives a scientific utility score based on freshness, redundancy, anomaly likelihood, and model uncertainty. The UAV then selects and visits a subset of high-utility cells under battery and return-to-base constraints. The proposed framework reframes UAV-based collection as adaptive scientific data management rather than exhaustive sensing.
Read the full abstract
Hardware Architecture
Interface-Aware KV Cache Quantization for Dense On-Chip NVM in Long-Context LLM Decoding
The key-value (KV) cache is the dominant memory bottleneck in long-context large language model (LLM) decoding: every step reads it entirely, so decoding is memory-bandwidth bound. Holding a quantized KV cache in dense on-chip non-volatile memory (NVM) removes the off-chip transfer. Existing KV quantization methods, however, were designed for GPU-style memory systems: KIVI attaches per-group metadata, adding about 25% to the stored KV cache; KVQuant keeps sparse full-precision outliers that a dense array cannot hold in place. This paper examines what these structures cost when the KV cache resides in NVM behind fixed-range converters, and designs a quantization scheme matched to that interface. The architecture stores the quantized KV cache in dense on-chip NVM, uses a small static analog crossbar only for the fixed rotation, and keeps attention in on-chip digital logic. A randomized rotation and per-vector normalization give every coordinate the same range, so one fixed codebook for keys and one for values, each shared across all tokens of the corresponding tensor type, serve the entire KV cache. The codebook thresholds are programmed once as the read converter's reference levels, enabling fixed-range digitization with no per-token converter reconfiguration. Dequantization is a sixteen-entry lookup and one norm multiply; the only per-vector metadata is one scalar, about 3%. Across models from 3B to 14B and contexts to 32k tokens, the four-bit KV cache maintains accuracy under storage and crossbar noise simulated at realistic device levels. KIVI and KVQuant remain more accurate in software; the advantage of our format lies at the memory interface: 3.1-3.6x lower KV read energy than both mapped to the same NVM, and 8x lower metadata overhead than KIVI. The contribution is a KV quantization co-designed with the NVM memory interface rather than a new accuracy record.
Read the full abstract
Systems and Control
AgentHomeID - Agent-based modelling of building stock transformation: A multi-scale framework for policy assessment and infrastructure planning
Decarbonising the building sector is central to meeting climate targets, yet existing models rarely capture the interaction between system-level transformation dynamics and heterogeneous individual investment decisions. This work presents AgentHomeID, an agent-based model of building stock evolution in which owner behaviour, techno-economic constraints, and regulatory frameworks are represented explicitly at the level of individual buildings and their owners. The model differentiates owner-occupiers, private landlords, and institutional owners, using willingness-to-pay (WTP) parameters estimated from empirical decision-maker studies, and operates on both representative building archetypes and real building data derived from geographic information systems (GIS). We demonstrate this versatility across three applications. At national scale, scenario analysis for Germany to 2045 shows that removing binding renewable heating requirements substantially raises final energy demand even where envelope refurbishment is unchanged, and that subsidy allocation and investment activity diverge sharply across owner types and income quartiles, with the lowest quartiles persistently underinvesting. At regional scale, bottom-up simulation for a German distribution grid planning region yields spatially concentrated heat pump uptake at NUTS-3 level that differs from aggregated top-down projections in both magnitude and spatial distribution. At urban block level, the same simulations resolve substation-level load heterogeneity and show that integrated system peaks driven by heat pumps, electric vehicles, and photovoltaics do not coincide with individual technology peaks. Across all three scales, owner heterogeneity and local structure materially shape transition pathways, indicating that they should be represented explicitly in models used for policy assessment and infrastructure planning.
Read the full abstract
Computer Vision
AAMBERS-UAV: Acquisition-Aware Multimodal Backbone Evaluation and Ranking for UAV Weedy Rice Segmentation
UAV image collections contain spatially and temporally related frames, yet semantic-segmentation benchmarks commonly split them at image level. Such splitting can place samples from one acquisition in both model development and testing, obscuring transfer to a genuinely new survey. Using the 734-sample WeedyRice-RGBMS-DB, we fix a 124-image target-acquisition test set and compare two protocols with identical train, validation, and test counts: target-held-out, which excludes the target acquisition from development, and target-exposed, which admits its remaining images. SegFormer-B0 is evaluated with RGB, four-band multispectral (MS), and seven-channel RGB+MS input over two fixed-split seeds. RGB is strongest under complete acquisition holdout ($0.7317\pm0.0201$ IoU), whereas RGB+MS becomes strongest after target exposure ($0.7822\pm0.0269$). A fixed-split U-Net/ResNet18 replication confirms positive exposure gains for all three inputs, but retains RGB as the best modality under both protocols. Acquisition exposure therefore increases measured performance across both evaluated backbones, while its effect on modality ranking is architecture-dependent. A supplied-split audit reveals strong near-sequential dependence, and corruption tests show that early fusion is substantially more sensitive to RGB–MS displacement than to moderate radiometric scaling. These results support acquisition-aware same-test evaluation as a necessary complement to ordinary image-level splitting in multimodal UAV benchmarks. The code and supporting the findings of this study will be publicly released upon acceptance of the paper.
Read the full abstract
Computer Vision
GeoContext: One Context Ladder, Two Failure Modes in Vision-Language Geolocation: Flat Reliance on User-Provided Location Context and False Confirmation of Location Claims
Visual geolocation benchmarks typically ask a model where an image was captured without accounting for the location context that users often provide. We introduce GeoContext, a resource supporting two complementary tasks: GeoHint, open-ended localization given a true but coarse location hint, and GeoVerify, binary verification of whether an image was taken within 150 m of a claimed place. GeoContext constructs a context ladder by stratifying nearby reference points according to distance and referenceability, allowing the image to remain fixed while the supplied context varies. The benchmark covers 109 sites in 30 cities and evaluates five vision-language models using 21,933 GeoHint responses and 6,270 GeoVerify responses. Our evaluation reveals three main patterns. First, hint repetition varies by only 1.5 percentage points across referenceability tiers and by less than 3 points across distance bands, while the resulting localization error increases steadily with hint distance. Second, behavior depends strongly on no-context performance: at sites with low no-context accuracy, the median ratio between localization error and hint distance is approximately 1.00, whereas at higher-accuracy sites it ranges from 0.24 to 0.69. After correcting for bias introduced by the site grouping procedure, only one of the five models retains a negative accuracy estimate when given a nearby hint. Third, in GeoVerify, no model reaches d' = 1 for decoys immediately beyond the 150 m tolerance. Model rankings also change when sensitivity is separated from response bias, and 83.8% of false acceptances are reported with confidence of at least 0.8. We release the benchmark, construction pipeline, audit decisions, and scoring code.
Read the full abstract
Performance
RAGMark: A Comprehensive Framework for Benchmarking Retrieval-Augmented Generation Systems
We present RAGMark, a modular benchmarking framework for advanced Retrieval-Augmented Generation (RAG) systems targeting small-scale multi-GPU environments. RAGMark evaluates diverse RAG components, including retrievers, vector databases, prompt-processing methods, and generator models, while collecting detailed per-stage metrics such as latency, GPU utilization, memory consumption, power usage, time to first token (TTFT), throughput, and answer quality. The framework is highly extensible, separating RAG stages, timing, and resource monitoring into modular components, and is designed to efficiently sweep large configuration spaces while minimizing repeated model and database initialization overhead. Using RAGMark, we characterize five RAG workloads on open-domain QA datasets across varying retrieval depths, model scales, reranking, compression methods, and vector database configurations. We show that while autoregressive generation dominates latency in naive pipelines, context-reduction techniques shift bottlenecks across compute, memory bandwidth, and preprocessing stages. Reranking and compression produce compounding benefits: reranking reduces compression workload itself, while both jointly reduce prefill and KV-cache traversal costs, lowering energy consumption by up to 66%. We further observe strong cross-stage interactions, where small upstream context reductions cascade through downstream latency, memory traffic, and energy consumption. The RAGMark source code is publicly available at: https://github.com/zferic/RAGMark.
Read the full abstract
Information Retrieval
What Price Fairness? Evaluating Energy - Fairness - Accuracy Trade-off in Recommender Systems
Fairness-aware recommender systems aim to mitigate systematic imbalances in recommendation outcomes, including how visibility, relevance, and opportunities are distributed among users, items, and providers. However, these systems are usually evaluated in terms of accuracy and fairness alone, while their computational and environmental costs remain largely invisible. This omission matters because fairness interventions may affect the cost of recommendation in different ways. Training-time methods modify model optimization, post-processing methods add computation at inference time, and both may depend on the model, dataset, hardware, and deployment setting. We examine whether provider-side fairness in recommendation comes with a measurable green cost. We compare in-processing, graph-level reweighting and post-processing interventions across multiple models, two datasets, and two hardware settings. We measure recommendation quality, provider-side exposure, and energy consumption separately across training and inference stages. Our results show that the green cost of fairness is not uniform, post-processing shifts cost to repeated serving, while in-processing and graph-level methods avoid re-ranking overhead but vary substantially across models, datasets, and hardware. Findings call for evaluating fairness-aware recommendation as a three-way trade-off between accuracy, fairness, and computational cost.
Read the full abstract
Artificial Intelligence
From Monolithic Blending to Agentic Orchestration: Dynamic Response for Conversational Assistants at Scale
Conversational assistants can blend retrieval, action selection, escalation, and wording in a single model path, or separate those roles. We report a production migration of a customer-support assistant at a large accommodation marketplace (millions of conversations per month, 11 languages, 10-second P90). Dynamic Response (DR) replaces a single Qwen3-235B-A22B blended responder with a bounded ReAct orchestrator over typed tools plus a smaller generator that writes from a backend-validated context contract. Because the migration also changed prompts, alignment, and serving, we attribute each effect to its cause and claim as architecture effects only those measured on identical replayed turns: typed entity selection moves the reservation selector to a precision-first operating point (precision 8.3% to 89.1%, recall 75.2% to 67.3%), and typed action IDs with a membership check remove observed structured-action hallucination (2.14% to 0.0%). A low-ramp A/B test reproduces the replay escalation reductions: hard-escalation responses fall from 5.60% to 3.08% and soft-escalation responses from 9.56% to 2.49%, while production handoff volume holds roughly steady; self-solve is directional (+5.1 points, 95% CI [-2, +12]). Serving optimizations cut orchestrator P90 latency from 3.87s to 2.24s on a GPU footprint reduced by roughly one-third, and self-hosting reduces estimated annual model-serving cost by more than an order of magnitude.
Read the full abstract
Computation and Language
CrisisKD: Five-Stage Knowledge Distillation for Aspect-Level Sentiment and Emotion Analysis in Crisis Discourse
Identifying the target of emotional words or phrases in crisis situations, especially health-related ones, is important for understanding public concerns across cultural and linguistic contexts. We propose CrisisKD, a five-stage teacher–student knowledge distillation framework for aspect-level sentiment and emotion analysis on unannotated social media data. A teacher LLM generates aspect-level labels and reasoning traces that supervise a smaller student model across aspect extraction, syntactic parsing, opinion extraction, sentiment classification, and emotion classification. Using this framework, we construct and release a dataset containing 50,615 aspect-level labels, together with the annotation and fine-tuning scripts as open-source resources. The resulting student supports end-to-end ABSA and emotion detection at substantially lower inference cost than the teacher. On a manually annotated 500-tweet gold set, the 5-task Qwen2.5-7B student improves over the untuned model by 7.9 F1 points on aspect extraction, 17.0 points on emotion accuracy, and 6.5 points on sentiment accuracy. On the external ABEA benchmark, CrisisKD improves the same-model Qwen2.5-7B ICL baseline by 2.8 F1 points on ATE and 3.8 F1 points on joint ATE+AEC.
Read the full abstract
Databases
Concord: A Video Relational Algebra for Cross-Modal Query Optimization
Semantic video queries let users embed natural language prompts and use multimodal large language models (MLLMs) to interpret the video. Such queries are increasingly popular for querying video data. However, their expressiveness comes at a steep cost: an MLLM may process hours of media to return only seconds of relevant output, making naive execution slow, expensive, and inaccurate. We propose Concord, a system for expressing and optimizing semantic video queries. We makes three contributions. First, we introduce Video Relational Algebra (VRA), a nested algebra over videos, transcripts, frames, and object tracks that captures common semantic video operations. Second, we derive a set of approximate optimizations that rewrite VRA queries to reduce MLLM usage while improving result quality. For narrated video, Concord either processes transcripts instead of video or uses them to identify video clips for MLLM processing. For cross-camera queries without narration, detection and tracking replace a whole-video MLLM join with a track-level relational join. Third, we evaluate Concord on real-world videos. Across 4.59 hours of soccer broadcasts and 3.92 hours of lectures, transcript-to-video queries send only 5.32% and 2.47% of source-video duration to the MLLM and reduce MLLM cost by up to 87%. In two five-second highway clips with 18 manually adjudicated cross-camera vehicles, a Detect-Track-Join query improves F1 from .364 to .813 while making no MLLM calls. See our project at https://concord-db.github.io/.