Silver Bullet

For art historians, goldsmith marks play a critical role in the identification and dating of artifacts. In practice, experts must manually compare a query mark against hundreds of documented examples, a process that is both tedious and highly dependent on specialist knowledge. To address this, we present an AI-assisted retrieval pipeline that combines mark localization with metric-learning fine-tuning across three backbone architectures: an ImageNet-pretrained ResNet-50, a supervised ViT-S/16, and a self-supervised DINOv2 ViT-S/14. We conduct a systematic evaluation of cropping strategies, where we measure the impact of no cropping, manual ground-truth cropping, and learned detection-based cropping, and assess their interaction with each backbone. Our strongest configuration, DINOv2 ViT-S/14 with manual crop and metric-learning fine-tuning, achieves an mAP of 62.63% and a Top-1 accuracy of 73.74%. Our experiments show that self-supervised pretraining and mark localization are the two most impactful factors, with learned cropping recovering the majority of the gain from manual cropping without requiring ground-truth annotations at inference time. To enable reproducibility and adoption in the digital humanities, we release our manually annotated dataset and codebase, and deploy the system via a public web interface.

Virtual try-on (VTON) enables customers to visualize how fashion products appear when worn and has become an important technology for online shopping. While recent advances have substantially improved garment VTON, jewelry remains a challenging and underexplored category due to its small size, rigid structure, and sensitivity to fine-grained visual details. Realistic jewelry VTON requires not only faithful appearance transfer but also accurate scale and placement relative to the wearer. Existing jewelry VTON methods typically rely on mask guidance, whereas mask-free approaches lack explicit guidance for modeling the product scale. To bridge this gap, we introduce JVTO-Bench, a benchmark dataset for scale-faithful jewelry VTON, providing reference source target triplets with real-world product-scale annotations across four major jewelry categories. Building upon this benchmark, we propose JewelTry, a mask-free diffusion framework for scale-aware jewelry VTON. JewelTry incorporates a scale adapter that encodes product dimensions into a scale token, enabling the model to learn scale relationships between jewelry items and surrounding human anatomy in-context. To further improve jewelry consistency, we introduce a single-directional condition attention mechanism and an attention refinement loss that preserve both coarse geometry and fine-grained structural details of the reference jewelry. Extensive experiments show that JewelTry achieves a balance among visual fidelity, background preservation, object consistency and scale accuracy, establishing a strong baseline for mask-free, scale-aware jewelry virtual try-on.

A correct answer can conceal why an agent succeeded. Once agents change their information state during evaluation, correctness no longer distinguishes intended reasoning from answer acquisition. Outcome evidence and exposure detection do not establish whether success depended on an acquired target; we call this missing evaluation object success provenance. AcquaBench audits it through matched CLEAN, GOLD, and SHAM value substitution on four standardized surfaces with joint qid-clustered analysis. CLEAN retains benchmark-authorized information. GOLD makes the correct target available. SHAM preserves source structure and exposure opportunity but substitutes a matched incorrect value. GOLD minus CLEAN measures the total score response to correct-target availability; GOLD minus SHAM tests whether that response tracks target correctness beyond matched source exposure. In D0, GOLD exceeds SHAM by 19.1 to 25.9 percentage points, showing that success follows the correct value. In D2, GOLD still exceeds SHAM under distributed sufficiency while coloc no longer transfers as a high-score marker, with AUROC 0.376 and 0.142. Behavioral dependence can thus persist beyond this probe's intended observation unit. In model comparison, a supported 5.0-point CLEAN score gap compresses to a raw GOLD difference of -0.6 points without establishing rank inversion. Agent benchmarks should report success together with whether the evaluated information state supported it.

On-device research agents search a corpus, read sources, and write a cited brief on a personal laptop. Whether their citations are faithful, and at what cost, is unmeasured for a deployable small model. This study fixes one 4B generator on a 24 GB laptop and asks what makes its citations faithful. It separates two quantities usually reported as one number. Cited claim faithfulness asks whether the cited source supports the claim. Trustworthy coverage asks whether the agent also cites the right sources. The study crosses how much of each source the generator sees, 400 against 1500 characters, with the quality of the sources supplied, gold papers against retrieved papers. Two levers fall out, and they act on different outcomes. Exposure sets faithfulness. More of each source lifts faithfulness from 0.45 to 0.58 on retrieved sources and from 0.37 to 0.58 on gold sources, and the two settings converge, so faithfulness is bound by exposure, not by whether the source is correct. The exposure lift is robust to a second, independent judge; the exact convergence is tight under the primary judge and only approximate under the second. Retrieval sets coverage. Trustworthy coverage stays near 0.22 on retrieved sources at any exposure, because recall is held near 0.40, so exposure cannot fix which sources are cited. The extra exposure costs about 235 output tokens. The practical recipe is to raise per source exposure first, cheaply, and then treat retrieval recall as the only remaining lever.

Decades of orbital missions have produced multi-modal remote sensing data for the Moon, spanning optical imagery, spectroscopy, thermal emission, radar, gravity, and elemental composition. Yet these datasets remain fragmented across archives, and no benchmark exists for evaluating machine learning on lunar data. We introduce Moonstone, the first multi-modal foundation model benchmark for lunar remote sensing. Our contributions are: (1) a 28-channel, 128 pixels-per-degree (~237 m) global lunar pretraining dataset from seven instrument families across five missions, (2) MG-MAE, a modality-grouped masked autoencoder with per-group convolutional tokenizers, a shared Vision Transformer encoder, attention masking for missing modalities, coverage-adaptive masking for heterogeneous spatial coverage, and spectral continuity regularization for physically plausible reconstructions, and (3) a benchmark of six downstream tasks covering classification, regression, and segmentation. MG-MAE pretrained features outperform scratch baselines on all tasks and surpass both ImageNet-pretrained and vanilla MAE baselines by large margins. Data and code are available at https://huggingface.co/datasets/ayushprd/Moonstone and https://github.com/ayushprd/Moonstone .

Reliable generative AI models critically rely on expert human annotations to evaluate output quality, yet these "gold" labels are expensive to collect and limited in quantity. Organizations thus often turn to collecting vast but noisy "silver" labels from crowdsourced workers or vendor annotators as proxies for gold labels. Because gold remains the evaluation target, naively aggregating noisy silver labels may introduce bias, and estimators built on sparsely observed gold labels may have high variance to resolve the model performance gaps that guide practical decisions. Model evaluation has become an ongoing operational practice rather than a one-time exercise, with evaluation rounds repeating across model versions, releases, and content domains. A natural question is whether the previous historical evaluation data can be used to improve each new round of evaluation. We introduce HERO (History Enhanced RObust model evaluation), a novel framework that uses historical data to suppress bias (improve reliability) and reduce variance (improve sensitivity) in model performance evaluation. HERO calibrates silver labelers' performance learned from historical gold annotations, and stabilizes the resulting estimator by anchoring it to covariate information measured with high precision in the historical data. HERO can be broadly applied across multiple common evaluation tasks, and remains valid when only a subset of historical labelers appears in the current round. We establish conditions under which the bias and variance reductions hold, showcase HERO's performance in simulation studies, and demonstrate its effectiveness on real-world model evaluation benchmarking datasets.

This paper presents an innovative approach that enables the users to capture their hand and try the jewel ring on their hand. The user captures the image of the hand using the React Native base GUI of the mobile application and selects the ring that the user wants to try, and the output image will have the user's hand with the ring image. This approach is implemented using a combination of MediaPipe hand point detection and YOLO-V8 custom object detection. The hand image uploaded by the user first undergoes mediapipe hand point detection. It will give the hand points and a Region of Interest mask where the ring is going to be placed. Then the ring is passed through YOLO object detection, in which ring points are detected, and background is removed. After that, using vector algebra, the angular discrepancy between the finger's reference axis and the ring's principal axis is computed. Also, ring size is rescaled according to finger thickness, preserving the aspect ratio to maintain perceptual realism. Then the ring is placed on the hand image and the output image is generated and shown on the user screen.

A classical intuition holds that verifying a solution is easier than producing one. For today's coding agents, this intuition is being inverted: as foundation models develop stronger reasoning capabilities and engineering harnesses grow more sophisticated, generating complex candidate solutions is no longer difficult -- reliably verifying them has become the harder problem. Every verifier we can build is only a proxy for human intent, never the intent itself. This makes verification subject to a twofold difficulty: first, intent is underspecified by nature, making it inherently hard to faithfully check whether it has been fulfilled; second, during model training, optimization widens the gap between proxy and intent -- manifesting as reward hacking or signal saturation. To address this, we characterize the quality of verification signals along three dimensions -- scalability, faithfulness, and robustness -- and argue that achieving all three simultaneously is the central challenge. We further study four reward constructions: a test verifier for general coding tasks, a rubric verifier for frontend tasks, the user as verifier for real-world agent tasks, and an automated agent verifier for long-horizon tasks. Across different task types and policy capability levels, we conduct in-depth analysis and experiments on the core challenges of reward design and how to more effectively leverage reward signals. Experiments show that targeted verification design can effectively suppress reward hacking, improve task completion quality, and achieve significant gains across multiple internal and public benchmarks. These experiences collectively point to a core observation: no fixed reward function can remain effective as policy capability continues to grow; and verification must co-evolve with the generator.

Scientific long-document summarization datasets commonly treat author-written abstracts as gold reference summaries, although their quality and alignment with the source article vary. At the same time, publicly available scientific summarization datasets remain limited in scale and structure for modern long-context models. In this work, we address both challenges by a) constructing and releasing one of the largest biomedical and life science datasets for long-document summarization, containing 1.88 million PMC articles, and b) analyzing the reference quality of author-written abstracts with source-grounded and model-based metrics. We show that author-written abstracts vary in their alignment with the full article and that these quality signals can guide training-data selection. Training on selected high-quality subsets outperforms random sampling at matched training sizes and can match or exceed larger random subsets on factuality-oriented metrics. Our findings suggest that reference quality is an important factor in scientific summarization and that quality-aware data selection can improve training efficiency.

This paper investigates the algebraic structure of Krom logic programs, consisting only of facts and rules with at most one body atom. We show that sequential composition endows the class of Krom programs with a natural monoid structure and that this structure admits rich algebraic extensions to Krom seminearrings, Krom quemirings, Krom-Conway seminearrings, and Krom-Conway omegaseminearrings. Furthermore, we establish explicit generating sets and canonical decompositions, study the associated ${}^ω$-operator, characterize the Kleene star in graph-theoretic terms, and relate finite Krom monoids to transformation monoids and finite-state automata. These results provide new connections between logic programming, algebraic automata theory, and algebraic graph theory.

People often see animal shapes in clouds, a phenomenon known as pareidolia. We propose an AI-based method that aims to predict which animals people are likely to perceive in clouds, even though state-of-the-art recognition methods typically fail to detect such animals. Additionally, we introduce a method to assist individuals in perceiving specific pareidolic animals, even if they did not recognize them initially. Our approach uses a diffusion model to transform cloud segments into an animal shape that visually resemble the original cloud. This diffusion technique is inspired by the observation that the diffusion process succeeds only when the target animal resembles the shape of the cloud, and that subtle visual hints often suffice to help individuals recognize specific pareidolic animals. A generated image, successfully derived from the diffusion model, is then used to predict the pareidolic animal. Additionally, a short morphing video transitioning from the generated image back to the original cloud segment is employed to further enhance the human's perception of the pareidolic animals.

The grading of gemstones is currently a manual procedure performed by gemologists. A popular approach uses reference stones, where those are visually inspected by specialists that decide which one of the available reference stone is the most similar to the inspected stone. This procedure is very subjective as different specialists may end up with different grading choices. This work proposes a complete framework that entails the image acquisition and goes up to the final stone categorization. The proposal is able to automate the entire process apart from including the stone in the created chamber for the image acquisition. It discards the subjective decisions made by specialists. This is the first work to propose a machine learning approach coupled with image processing techniques for emerald grading. The proposed framework achieves 98% of accuracy (correctly categorized stones), outperforming a deep learning approach. Furthermore, we also create and publish the used dataset that contains 192 images of emerald stones along with their extracted and pre-processed features.

Modelling claim frequency and severity for non-life insurance pricing predominantly relies on generalised linear models, with gradient-boosted machines as the leading machine learning alternative. Tabular foundation models (TFMs) present a fundamentally different inference paradigm. By pre-training on large collections of synthetic datasets, TFMs enable inference on new data through in-context learning, without any dataset-specific fitting or hyperparameter tuning. This paper presents a first empirical evaluation of TabPFN for motor insurance pricing, benchmarking it against GLM and XGBoost on two publicly available MTPL datasets. Our results show that TabPFN does not consistently outperform established baselines, exhibits substantially longer inference times, and is sensitive to the size of the in-context training set. While tabular foundation models represent a promising direction, particularly in data-scarce settings, their current performance does not offer a viable replacement for established actuarial methods.

This paper investigates chunking strategies for retrieval-augmented generation on German statutory law, using the German Civil Code as a structured benchmark corpus. We implement and compare a range of segmentation approaches, including structural units (sections, subsections, sentences, propositions), fixed-size windows, contextual chunking, semantic clustering, Lumber-style chunking, and RAPTOR-based hierarchical retrieval. All methods are evaluated on a legal question-answering dataset with section-level gold labels, measuring recall, query latency, index build time, and storage requirements. Results show that chunking strategies aligned with the inherent legal structure - particularly section and subsection - based retrieval-achieve the highest recall, while more complex approaches that override this structure perform worse. These simpler methods also offer favorable computational efficiency compared to LLM-intensive techniques such as contextual chunking, RAPTOR, and Lumber. The findings highlight a key trade-off between semantic enrichment and operational cost, and demonstrate that preserving domain-specific structure is critical for effective legal information retrieval.

Extreme low-data fine-grained classification is common in expert domains where labeling is expensive, yet practitioners still need principled guidance for selecting pretrained encoders. We study emerald inclusion grading with a custom dataset of labeled images across three classes and ask: under matched backbone capacity, how does pretraining objective affect downstream representation quality? We compare four frozen ViT-B/16 encoders trained with supervised classification, contrastive learning (SigLIP2), masked reconstruction (MAE), and self-distillation (DINOv3), and evaluate them with leave-one-out cross-validation using linear and nonlinear probes. To control statistical noise in the low-N regime, we use permutation testing (N=1000) on macro one-vs-rest AUC. Supervised and contrastive encoders provide the strongest linear separability (logistic AUC: 0.768 and 0.735; SVM AUC: 0.739 and 0.697), while MAE improves under nonlinear probes (XGBoost AUC: 0.713). We find that DINOv3 underperforms across probe families in this domain. These results support a practical recommendation for extreme low-data FGVC: prioritize margin-enforcing pretraining objectives when data scarcity restricts probing to linear decision rules, and consider reconstruction-style encoders when nonlinear classifiers are feasible given dataset constraints.

In physical systems, whenever a continuous symmetry is spontaneously broken, the system possesses excitations called Goldstone modes, which allow coherent information propagation over long distances and times. In this work, we study deep neural networks whose internal layers are equivariant under a continuous symmetry and may therefore support analogous Goldstone-like degrees of freedom. We demonstrate, both analytically and empirically, that these degrees of freedom enable coherent signal propagation across depth and recurrent iterations, providing a mechanism for stable information flow without relying on architectural stabilizers such as residual connections or normalization. In feedforward networks, this results in improved trainability and representational diversity across layers. In recurrent settings, we demonstrate the same mechanism is valuable for long-term memory by propagating information over recurrent iterations, thereby improving performance of RNNs and GRUs on long-sequence modeling tasks.

Generative AI can encode substantial medical knowledge, but patient-specific answers remain constrained by the context supplied at inference time. Electronic health records and FHIR support documentation and interoperability, but they do not by themselves define the complete, current, and auditable context a model should receive. Similarity-based retrieval can find related text, but it neither guarantees collection of clinically connected records nor makes omissions explicit. We introduce MedBeads, an AI-facing clinical record substrate that assembles a declared closure of longitudinal patient information before generation. A Bead is an immutable clinical or knowledge object identified by SHA-256 over canonical content and stored as an append-only frame in a patient-scoped Pod. Structural parent edges form a patient-rooted Merkle DAG. Typed clinical links occupy a separate, reconstructable interpretation layer derived from signed, versioned knowledge rules; they can be recomputed when knowledge changes without rewriting clinical facts. Retrieval follows authorized structural and clinical edges, resolves amendments and retractions, and reports policy or token truncation. An open-source Go implementation uses append-only Pods and reconstructable SQLite projections. File-based conversion of 1,135 synthetic Synthea FHIR bundles produced approximately one million Beads and demonstrated deterministic clinical-link derivation and interpretation-layer reconstruction. These engineering results establish feasibility and reproducibility, not reduced hallucination or improved clinical outcomes. MedBeads reframes grounding as a data-structure problem by delivering a policy-bounded, provenance-bearing clinical subgraph rather than an opaque list of similar fragments.

Large-scale models are pretrained on massive web-crawled datasets containing documents of mixed quality, making data filtering essential. A popular method is Classifier-based Quality Filtering (CQF), which trains a binary classifier to distinguish between pretraining data and a small, high-quality set. It assigns each pretraining document a quality score defined as the classifier's score and retains only the top-scoring ones. We provide an in-depth analysis of CQF. We show that while CQF improves downstream task performance, it does not necessarily enhance language modeling on the high-quality set. Importantly, we find that training on CQF-selected data can outperform training directly on the high-quality set, even when the latter is sufficiently large. This finding alone is particularly striking, given the substantial effort and cost recently devoted to augmenting high-quality data. We explain this paradox by the fact that CQF implicitly filters the high-quality dataset as well as the low-quality one. Finally, we introduce an optimization-driven notion of data quality and demonstrate that it can be reliably estimated using small-scale proxy experiments. Altogether, our results both elucidate the mechanisms behind CQF and deepen our understanding of data selection methods widely used in practice.

We study how far gradient descent (GD) can be accelerated by predetermined stepsizes in smooth convex optimization. Writing $p_{\mathrm{sil}}=\log_2(1+\sqrt{2})$, we prove an $\Omega\left(n^{-p_{\mathrm{sil}}-O(\sqrt{\log\log n/\log n})}\right)$ non-anytime lower bound. In the anytime setting, every infinite schedule has infinitely many horizons with error $\Omega\left(n^{-\frac{2p_{\mathrm{sil}}}{1+p_{\mathrm{sil}}}-O(\sqrt{\log\log n/\log n})}\right)$. Together with the silver-schedule upper bound [Altschuler and Parrilo, 2025] and the anytime upper bound [Zhang et al., 2025], our results determine the optimal polynomial convergence exponents in both settings.

Knowledge graph question answering seeks to translate natural language questions into executable queries over knowledge graphs, but existing approaches often rely on large models or full supervision in the form of gold query annotations. This study examines whether reinforcement learning with outcome-based rewards can train a small instruction-tuned language model to perform zero-shot Text-to-SPARQL generation in the scholarly domain. Group-Relative Policy Optimization (GRPO) is applied to the Qwen3-1.7B model on DBLP-QuAD, using prompts that combine natural language questions with symbolic hints about entities and relations. Training relies on execution feedback, structural constraints, and answer-level rewards, with an additional variant that incorporates gold-query-based shaping. The resulting models are compared to the unmodified zero-shot baseline and to a supervised DoRA-finetuned baseline across answer-level accuracy, execution accuracy, category-wise scores, and generalization to held-out templates. GRPO substantially improves over the zero-shot baseline and exhibits competitive generalization, while supervised DoRA finetuning achieves higher overall accuracy on the same model scale. Ablation analyses indicate that execution-based rewards account for most gains, with additional shaping yielding limited additional benefit, suggesting that outcome-based reinforcement learning is a viable training strategy when gold queries are unavailable for token-level supervision.