Abstract

Knowledge Graph Foundation Models (KGFMs) have shown promise in enabling zero-shot reasoning over unseen graphs by learning transferable patterns. However, most existing KGFMs rely solely on graph structure, overlooking the rich semantic signals encoded in textual attributes. We introduce SEMMA, a dual-module KGFM that systematically integrates transferable textual semantics alongside structure. SEMMA leverages Large Language Models (LLMs) to enrich relation identifiers, generating semantic embeddings that subsequently form a textual relation graph, which is fused with the structural component. Across 54 diverse KGs, SEMMA outperforms purely structural baselines like ULTRA in fully inductive link prediction. Crucially, we show that in more challenging generalization settings, where the test-time relation vocabulary is entirely unseen, structural methods collapse while SEMMA is 2x more effective. Our findings demonstrate that textual semantics are critical for generalization in settings where structure alone fails, highlighting the need for foundation models that unify structural and linguistic signals in knowledge reasoning.

Abstract

This paper presents an 18.87–24.19 GHz dual-core, dual-mode voltage-controlled oscillator (VCO) using a multi-tap (MT) inductor-based 2-port resonator with capacitive coupling via a mode-switching block. The 2-port resonator introduces a gate-to-drain phase shift, suppressing flicker phase noise (PN), while the high-quality factor of the tank minimizes thermal PN. The mode-switching block enables dual resonance modes—high-frequency band (HB) and low-frequency band (LB), achieving a wide frequency tuning range (FTR) and low PN in both the 1/f^2 and 1/f^3 regions. The proposed VCO is implemented in 65 nm CMOS, and post-layout results show a PN of –68.7 dBc/Hz at a 10 kHz offset and –118.2 dBc/Hz at a 1 MHz offset from an 18.87 GHz carrier frequency, with a power consumption of 25.65 mW and an area of 0.154 mm^2. The peak figure of merit (FoM) is 189.6 dBc/Hz, and the maximum 1/f^3 noise corner frequency is 347 kHz across the FTR.

Abstract

We explore language-agnostic deep text embeddings for severity classification of dysarthria in Amyotrophic Lateral Sclerosis (ALS). Speech recordings are transcribed by human and ASR and embeddings of the transcripts are considered. Though speech recognition accuracy has been studied for grading dysarthria severity, no effort has yet been made to utilize text embeddings of the transcripts. We perform severity classification at different granularity (2, 3, and 5-class) using data obtained from 47 ALS subjects. Experiments with dense neural network based classifiers suggest that, though text features achieve nearly equal performances as baseline speech features, like statistics of mel frequency cepstral coefficients (MFCC), for 2-class classification, speech features outperform for higher number of classes. Concatenation of text embeddings and MFCC statistics attains the best performances with mean F1 scores of 88%, 68%, and 53%, respectively, in 2, 3, and 5-class classification.

Abstract

As robotic manipulators start performing more daily tasks, striking can be a useful method for transporting objects because it significantly increases the reachable workspace. However, striking methods are underexplored compared to pick-and-place because of the difficulty of modeling and executing striking interactions. In this paper, we develop an algorithm for striking objects so that they stop at a target location. We start with an optimizer in simulation that solves for the striking velocity given the relative target position, and perform system identification to set the simulation parameters. However, real-world striking with this model does not have high accuracy because it is unclear which parameters should be considered for identification in practice. Therefore, we finally developed a residual learning approach that subsumes all unmodeled differences between the simulation and the real-world environment into action, that is, striking velocity residuals. Our real-world experiments show that the residual learning model results in 81.6% more accurate object strikes.

Abstract

Following up on our earlier study [J. Bardhan et al., Phys. Rev. D 107, 115001 (2023)], we investigate the LHC prospects of pair-produced vectorlike 𝐵 quarks decaying exotically to a new gauge-singlet (pseudo)scalar field Φ and a 𝑏 quark. After the electroweak symmetry breaking, the Φ decays predominantly to 𝑔⁡𝑔⁡/𝑏⁢𝑏 final states, leading to a fully hadronic 2⁢𝑏 +4⁢𝑗 or 6⁢𝑏 signature. Because of the large Standard Model background and the lack of leptonic handles, it is a difficult channel to probe. To overcome the challenge, we employ a hybrid deep learning model containing a graph neural network followed by a deep neural network. We estimate that such a state-of-the-art deep learning analysis pipeline can lead to a performance comparable to that in the semileptonic mode, taking the discovery (exclusion) reach up to about 𝑀_𝐵=1.8⁢(2.4)  TeV at HL-LHC when 𝐵 decays fully exotically, i.e., BR⁡(𝐵→𝑏⁢Φ)=100%.

Abstract

This study advances Kai Katahira’s Speculation Game, an agent-based model (ABM) for financial markets, by addressing its limitation in capturing order flow imbalance, a critical indicator of herd behavior. Although the original model successfully replicated key stylized facts of financial markets, it did not account for the persistence of order imbalance observed in real-world trading. Through a comprehensive analysis of two decades of BSE Sensex data, we establish the prevalence of order imbalance and its correlation with price fluctuations. To bridge this gap, we propose an extended model, Speculation Game with Information Cascade (SGIC), which integrates Self-Organized Criticality (SOC) through a sand-pile model, enabling agents to interact within a small-world network. Our proposed model not only reproduces the stylized facts captured by the original Speculation Game, but also successfully generates the additional stylized fact of order flow imbalance. These advances enhance the realism of ABMs in financial markets, providing deeper insights into the mechanisms driving herding and market fluctuations.

Abstract

The color gamut of visual scenes can encompass a multitude of hues, but how many distinct hues are individuals aware of? We examined this question in the context of texture perception, with a grid of stimuli composed of random colors chosen from color sets varying from two to six different hues or saturations. In a behavioral experiment, participants had to discriminate between differences in the number of hues present by identifying which of four color grids included a larger number of different colors. Color number discrimination was also assessed neurally, using electroencephalogram (EEG) frequency tagging, wherein a texture with an extra hue was shown as an “oddball,” once per second, in a stream of textures presented at six images per second. In both experiments, two versus three texture contrasts were readily distinguished, yet performance fell rapidly when the set size increased further. The results suggest that sensitivity to the density of the color gamut defining a texture is very restricted.

Abstract

Understanding user intent is essential for build- ing better human interaction agents, as it enables personalization, co-creation, and contextual adap- tation. However, existing approaches are either restricted to text environments, use human anno- tation, or just predict future user actions lacking the ability to reason explicitly about user goals. In this work, we introduce EARL (Early Action Reasoning for Latent intent), a theory of mind inspired inference-time algorithm that models user intent as an inverse planning problem, in- ferring latent goals from observed user actions. EARL hypothesizes potential user intent at mul- tiple stages during the course of task execution, enabling timely intervention and personalization. Evaluated on three diverse benchmarks namely Mind2Web, AiTz, and VideoGUI, and using two strong LLMs (Gemini-1.5-Pro and GPT-4o), we show that EARL consistently outperforms CoT- based LLM baselines in accurately deciphering user intent, especially under partial observations

Abstract

Centrality measures offer a mechanism to quantify various notions of the importance of vertices in a network. These measures find application in social network analysis, the study of disease spread, and the like. Since the networks of interest are usually large, parallelism has become a powerful tool for computing these measures. However, for some of these applications, it is crucial to study how these centrality values change when there are changes to the underlying network. As a result, parallel algorithms that examine how to update these values due to changes to the graph are gaining research interest. In this paper, we study the update of the percolation centrality measure in a dynamic graph. We present parallel algorithms to handle changes to the percolation value of a batch of vertices and the addition and deletion of a batch of edges to the graph. We leverage the graphs' structural properties to improve our algorithms' performance. To our knowledge, we are the first to propose such algorithms for percolation centrality. We implement and benchmark our dynamic algorithms on a server with two AMD EPYC CPUs and an Nvidia A100 GPU. Our experiments on a collection of real-world graphs indicate that our algorithms achieve speedups of 8.79x and 2.82x on CPU and GPU, respectively, for edge updates, and 309.44x and 18.71x on CPU and GPU, respectively, for vertex percolation updates over state-of-the-art static algorithms for a batch of 10000 edges and vertices, respectively.

Abstract

Efficient algorithms for graph connected components and minimum spanning tree (MST) computations are vital in diverse domains, from medical diagnostics to chip design. Yet, as graph sizes grow to billions of edges, existing approaches often fail due to limited device memory. This challenge demands innovative rethinking of both algorithmic design and implementation strategies. We address this need by presenting a novel Parallel Heterogeneous External Memory (PHEM) framework, which harnesses both multicore CPU and GPU resources to tackle large-scale graph processing. Our method minimizes communication overhead while maximizing computational throughput through efficient copy- computation strategies. Experiments on a variety of real-world and synthetic benchmarks demonstrate that our algorithms for connected components and minimum spanning tree in the PHEM model outperform current state-of-the-art solutions, significantly advancing the capability to process massive graphs.