搜索结果：rediscovering

Olfaction plays an important role in human perception, yet its subjective and ephemeral nature makes it difficult to articulate, compare, and share across individuals. Traditional practices like the Japanese incense game Genji-ko offer one way to structure olfactory experience through shared interpretation. In this work, we present Smell with Genji, an AI-mediated olfactory interaction system that reinterprets Genji-ko as a collaborative human-AI sensory experience. By integrating a game setup, a mobile application, and an LLM-powered co-smelling partner equipped with olfactory sensing and LLM-based conversation, the system invites participants to compare scents and construct Genji-mon patterns, fostering reflection through a dialogue that highlights the alignment and discrepancies between human and machine perception. This work illustrates how sensing-enabled AI can participate in olfactory experience alongside users, pointing toward new possibilities for AI-supported sensory interaction and reflection in HCI.

RGB-based 3D tasks, e.g., 3D detection, depth estimation, 3D keypoint estimation, still suffer from scarce, expensive annotations and a thin augmentation toolbox, since many image transforms, including rotations and warps, disrupt geometric consistency. While horizontal flipping and color jitter are standard, rigorous 3D rotation augmentation has surprisingly remained absent from RGB-based pipelines, largely due to the misconception that it requires scene depth or scene reconstruction. In this paper, we introduce 3DRot, a plug-and-play augmentation that rotates and mirrors images about the camera's optical center while synchronously updating RGB images, camera intrinsics, object poses, and 3D annotations to preserve projective geometry, achieving geometry-consistent rotations and reflections without relying on any scene depth. We first validate 3DRot on a classical RGB-based 3D task, monocular 3D detection. On SUN RGB-D, inserting 3DRot into a frozen DINO-X + Cube R-CNN pipeline raises $IoU_{3D}$ from 43.21 to 44.51, cuts rotation error (ROT) from 22.91$^\circ$ to 20.93$^\circ$, and boosts $mAP_{0.5}$ from 35.70 to 38.11; smaller but consistent gains appear on a cross-domain IN10 s

Vision Language Models (VLMs) excel at identifying and describing objects but often fail at spatial reasoning. We study why VLMs, such as LLaVA, underutilize spatial cues despite having positional encodings and spatially rich vision encoder features. Our analysis reveals a key imbalance: vision token embeddings have much larger norms than text tokens, suppressing LLM's position embedding. To expose this mechanism, we developed three interpretability tools: (1) the Position Sensitivity Index, which quantifies reliance on token order, (2) the Cross Modality Balance, which reveals attention head allocation patterns, and (3) a RoPE Sensitivity probe, which measures dependence on rotary positional embeddings. These tools uncover that vision tokens and system prompts dominate attention. We validated our mechanistic understanding through targeted interventions that predictably restore positional sensitivity. These findings reveal previously unknown failure modes in multimodal attention and demonstrate how interpretability analysis can guide principled improvements.

The accurate modeling of the mechanical behavior of rubber-like materials under multi-axial loading constitutes a long-standing challenge in hyperelastic material modeling. This work employs deep symbolic regression as an interpretable machine learning approach to discover novel strain energy functions directly from experimental results, with a specific focus on the classical Treloar and Kawabata data sets for vulcanized rubber. The proposed approach circumvents traditional human model selection biases by exploring possible functional forms of strain energy functions, expressed in terms of both the first and second principal invariants of the right Cauchy-Green tensor. The resulting models exhibit high predictive accuracy for various deformation modes, including uniaxial tension, pure shear, equal biaxial tension, and biaxial loading. This work underscores the potential of deep symbolic regression in advancing hyperelastic material modeling and highlights the importance of considering both invariants in capturing the complex behaviors of rubber-like materials.

This work explores using the physics-inspired AI Feynman symbolic regression algorithm to automatically rediscover a fundamental equation in astronomy -- the Equation of the Centre. Through the introduction of observational and inductive biases corresponding to the physical nature of the system through data preprocessing and search space restriction, AI Feynman was successful in recovering the first-order analytical form of this equation from lunar ephemerides data. However, this manual approach highlights a key limitation in its reliance on expert-driven coordinate system selection. We therefore propose an automated preprocessing extension to find the canonical coordinate system. Results demonstrate that targeted domain knowledge embedding enables symbolic regression to rediscover physical laws, but also highlight further challenges in constraining symbolic regression to derive physics equations when leveraging domain knowledge through tailored biases.

Routing is central to networking performance, including: (1) latency in anycast services and websites served from multiple locations,(2) networking expenses and throughput in multi-homed enterprises, (3) the ability to keep traffic domestic when considering data sovereignty. However, understanding and managing how routing affects these services is challenging. Operators use Traffic Engineering (TE) with BGP to optimize network performance, but what they get is the result of all BGP policies throughout the Internet, not just their local choices. Our paper proposes Fenrir, a new system to rediscover recurring routing results. Fenrir can discover changes in network routing, even when it happens multiple hops away from the observer. Fenrir also provides new methods to quantify the degree of routing change, and to identify routing "modes" that may reappear. Second, we show that Fenrir can be applied to many different problems: we use five instances of three different types of systems to illustrate the generalization: anycast catchments showing in a root DNS service, route optimization for two multi-homed enterprises, and website selection for two of the top-10 web services. Each type re

We investigate whether artificial intelligence can autonomously recover known structures of the Standard Model of particle physics using only experimental data and without theoretical inputs. By applying unsupervised machine learning techniques -- including data dimensionality reduction and clustering algorithms -- to intrinsic particle properties and decay modes, we uncover key organizational features of particle physics, such as the relative strength of different interactions and the difference between baryons and mesons. We also identify conserved quantities such as baryon number, strangeness and charm as well as the structure of isospin and the Eightfold Way multiplets. Our analysis then reveals that clustering can separate particles by interaction, flavor symmetries as well as quantum numbers. Additionally, we observe patterns consistent with Regge trajectories in baryon excitations. Our results demonstrate that machine learning can reproduce key aspects of the Standard Model directly from data, suggesting a promising path toward data-driven discovery in fundamental physics.

Forecast reconciliation has become a prominent topic in recent forecasting literature, with a primary distinction made between cross-sectional and temporal hierarchies. This work focuses on temporal hierarchies, such as aggregating monthly time series data to annual data. We explore the impact of various forecast reconciliation methods on temporally aggregated ARIMA models, thereby bridging the fields of hierarchical forecast reconciliation and temporal aggregation both theoretically and experimentally. Our paper is the first to theoretically examine the effects of temporal hierarchical forecast reconciliation, demonstrating that the optimal method aligns with a bottom-up aggregation approach. To assess the practical implications and performance of the reconciled forecasts, we conduct a series of simulation studies, confirming that the findings extend to more complex models. This result helps explain the strong performance of the bottom-up approach observed in many prior studies. Finally, we apply our methods to real data examples, where we observe similar results.

Assessing personality traits using large language models (LLMs) has emerged as an interesting and challenging area of research. While previous methods employ explicit questionnaires, often derived from the Big Five model of personality, we hypothesize that LLMs implicitly encode notions of personality when modeling next-token responses. To demonstrate this, we introduce a novel approach that uncovers latent personality dimensions in LLMs by applying singular value de-composition (SVD) to the log-probabilities of trait-descriptive adjectives. Our experiments show that LLMs "rediscover" core personality traits such as extraversion, agreeableness, conscientiousness, neuroticism, and openness without relying on direct questionnaire inputs, with the top-5 factors corresponding to Big Five traits explaining 74.3% of the variance in the latent space. Moreover, we can use the derived principal components to assess personality along the Big Five dimensions, and achieve improvements in average personality prediction accuracy of up to 5% over fine-tuned models, and up to 21% over direct LLM-based scoring techniques.

We introduce a novel orbit superposition method designed to reconstruct the stellar density structure, kinematics, and chemical abundance distribution of the entire Milky Way by leveraging 6D phase-space information from its resolved stellar populations, limited by the spatial coverage of APOGEE DR17.

This paper introduces the concept of uniform classification, which employs a unified threshold to classify all samples rather than adaptive threshold classifying each individual sample. We also propose the uniform classification accuracy as a metric to measure the model's performance in uniform classification. Furthermore, begin with a naive loss, we mathematically derive a loss function suitable for the uniform classification, which is the BCE function integrated with a unified bias. We demonstrate the unified threshold could be learned via the bias. The extensive experiments on six classification datasets and three feature extraction models show that, compared to the SoftMax loss, the models trained with the BCE loss not only exhibit higher uniform classification accuracy but also higher sample-wise classification accuracy. In addition, the learned bias from BCE loss is very close to the unified threshold used in the uniform classification. The features extracted by the models trained with BCE loss not only possess uniformity but also demonstrate better intra-class compactness and inter-class distinctiveness, yielding superior performance on open-set tasks such as face recognitio

The Mullins effect represents a softening phenomenon observed in rubber-like materials and soft biological tissues. It is usually accompanied by many other inelastic effects like for example residual strain and induced anisotropy. In spite of the long term research and many material models proposed in literature, accurate modeling and prediction of this complex phenomenon still remain a challenging task. In this work, we present a novel approach using deep symbolic regression (DSR) to generate material models describing the Mullins effect in the context of nearly incompressible hyperelastic materials. The two step framework first identifies a strain energy function describing the primary loading. Subsequently, a damage function characterizing the softening behavior under cyclic loading is identified. The efficiency of the proposed approach is demonstrated through benchmark tests using the generalized the Mooney-Rivlin and the Ogden-Roxburgh model. The generalizability and robustness of the presented framework are thoroughly studied. In addition, the proposed methodology is extensively validated on a temperature-dependent data set, which demonstrates its versatile and reliable perfo

Scientific discovery plays a pivotal role in advancing human society, and recent progress in large language models (LLMs) suggests their potential to accelerate this process. However, it remains unclear whether LLMs can autonomously generate novel and valid hypotheses in chemistry. In this work, we investigate whether LLMs can discover high-quality chemistry hypotheses given only a research background-comprising a question and/or a survey-without restriction on the domain of the question. We begin with the observation that hypothesis discovery is a seemingly intractable task. To address this, we propose a formal mathematical decomposition grounded in a fundamental assumption: that most chemistry hypotheses can be composed from a research background and a set of inspirations. This decomposition leads to three practical subtasks-retrieving inspirations, composing hypotheses with inspirations, and ranking hypotheses - which together constitute a sufficient set of subtasks for the overall scientific discovery task. We further develop an agentic LLM framework, MOOSE-Chem, that is a direct implementation of this mathematical decomposition. To evaluate this framework, we construct a bench

New data-driven methods have advanced the discovery of governing equations from observations, enabling parsimonious models for complex systems. Here, we 'rediscover' a shallow-water equation closely related to Korteweg--de Vries (KdV) using only video recordings of solitons in a simple flume. Two fundamentally different approaches -- weak-form sparse identification of nonlinear dynamics (WSINDy) and a novel Fourier-multiplier method -- recover the same PDE, demonstrating that the equation is inherent in the data and robust to the choice of method. Both identify the same terms with comparable magnitudes and errors. To validate the models, we solve the discovered equations forward in time and compare them with additional experimental cases that were not used in the discovery. Based on the results, we discuss absolute and cumulative errors, as well as the strengths and limitations of the two discovery approaches. Together, these results demonstrate the potential of equation discovery from everyday experiments ('GoPro physics') and highlight shallow-water waves as an ideal test bed for developing and benchmarking new methods.

In the recent past, a popular way of evaluating natural language understanding (NLU), was to consider a model's ability to perform natural language inference (NLI) tasks. In this paper, we investigate if NLI tasks, that are rarely used for LLM evaluation, can still be informative for evaluating LLMs. Focusing on five different NLI benchmarks across six models of different scales, we investigate if they are able to discriminate models of different size and quality and how their accuracies develop during training. Furthermore, we investigate the extent to which the softmax distributions of models align with human distributions in cases where statements are ambiguous or vague. Overall, our results paint a positive picture for the NLI tasks: we find that they are able to discriminate well between models at various stages of training, yet are not (all) saturated. Furthermore, we find that while the similarity of model distributions with human label distributions increases with scale, it is still much higher than the similarity between two populations of humans, making it a potentially interesting statistic to consider.

The present study puts forward a novel biographical knowledge graph (KG) on Prof. S. R. Ranganathan, one of the pioneering figures in the Library and Information Science (LIS) domain. It has been found that most of the relevant facts about Ranganathan exist in a variety of resources (e.g., books, essays, journal articles, websites, blogs, etc.), offering information in a fragmented and piecemeal way. With this dedicated KG (henceforth known as RKG), we hope to furnish a 360-degree view of his life and achievements. To the best of our knowledge, such a dedicated representation is unparalleled in its scope and coverage: using state-of-the-art technology for anyone to openly access, use/re-use, and contribute. Inspired by Ranganathan's theories and ideas, the KG was developed using a "facet-based methodology" at two levels: in the identification of the vital biographical aspects and the development of the ontological model. Finally, with this study, we call for a community-driven effort to enhance the KG and pay homage to the Father of Library Science on the hundredth anniversary of his revitalizing the LIS domain through his enduring participation.

Training and inference on edge devices often requires an efficient setup due to computational limitations. While pre-computing data representations and caching them on a server can mitigate extensive edge device computation, this leads to two challenges. First, the amount of storage required on the server that scales linearly with the number of instances. Second, the bandwidth required to send extensively large amounts of data to an edge device. To reduce the memory footprint of pre-computed data representations, we propose a simple, yet effective approach that uses randomly initialized hyperplane projections. To further reduce their size by up to 98.96%, we quantize the resulting floating-point representations into binary vectors. Despite the greatly reduced size, we show that the embeddings remain effective for training models across various English and German sentence classification tasks that retain 94%--99% of their floating-point.