Osaka University of Economics

Understanding animal actions and interactions is essential for behavior analysis and ecological monitoring. Although large-scale in-the-wild datasets have advanced animal action recognition, existing methods still struggle with fine-grained motion, spatial relations, and multi-individual interactions. To address these challenges, we introduce AIRA, a unified framework for Action and Interaction Recognition in Animals. Built upon a vision-language model (VLM), AIRA learns in an action-centered representation space defined by body parts and their corresponding motions, thereby improving robustness to background noise and enabling cross-species generalization via a unified mammal-centric part ontology. To model actions, we treat body parts and motion as primary cues and introduce Actionbit tokens-compact representations for parts and motions generated by a large language model (LLM) that encode which parts move and how. We further propose Part-Enhanced Prompt Fine-tuning (PEPF) to make the VLM explicitly sensitive to part and pose cues. Within PEPF, the Action-actionbit Alignment (AbA) module enriches action representations with fine-grained part-motion semantics, and Part-Vision Prompting (PVP) extracts keyframes through action-aware prompting. Experiments across multiple benchmarks show consistent improvements in both action and interaction recognition, highlighting the importance of action-centered adaptation and relational reasoning for understanding animal behavior in the wild.

Efficiency of optical beats in a chaotically oscillating laser is confirmed comparing that of free running CW laser using a highly efficient plasmonic photomixer. Ultra-stability of optical beats with laser chaos continues down to the laser threshold level. The output power is also estimated using conventional bowtie antenna. The great potential of chaotically oscillating lasers is verified for THz systems.

Reconstructing hand–object interactions from visual data is difficult under severe occlusions, especially when soft objects are involved. In this work, we explore the use of thermal imagery as an additional cue for refining RGBbased hand–object reconstruction. We map multi-view thermal observations onto object surfaces and represent delayed thermal traces of contact as a contact potential field. This field is incorporated as a soft temporal cue to refine hand–object relative poses. Preliminary results indicate that thermal guidance can help resolve visually ambiguous configurations caused by occlusions.

A series of ruthenium complexes showing unique solvatochromism in the chlorido‐ligated form and reversible photoisomerization in the aqua‐ligated form has been developed. The complexes bear a tridentate ligand and a bidentate ligand containing naphthyridine and diethylamino moieties. In the chlorido complex, steric repulsion between the chlorido ligand and the diethylamino group promotes ligand‐substitution with solvent molecules in coordinating and protic solvents. The absorption maxima in the visible region shifted by approximately 120 nm depending on the solvent. The aqua complex exhibited intramolecular hydrogen bonding between the aqua and diethylamino moieties according to X‐ray crystallography. The photoisomerization behavior of the aqua complex was evaluated by 1 H nuclear magnetic resonance measurements. The quantum yield of photoisomerization was 2 × 10 −4 , which is much smaller than that of the complex without the diethylamino moiety, indicating that the hydrogen bonding hinders the photodissociation of the aqua ligand.

Domestic squalor (DS) is related to physical problems, living alone, family and neighbourhood problems and mental illnesses, especially dementia. However, methods to detect and treat older adults with DS remain unclear. We aimed to clarify the methods of appropriate intervention for older adults with DS. The Initial-phase Intensive Support Team for Dementia (IPIST) is a multidisciplinary outreach team in Japan that provides intensive support to people living at home with suspected dementia. We distributed a questionnaire to 50 IPISTs, specifically requesting the submission of 'two complex cases that were difficult to manage' regarding medical or long-term care. The questionnaire included sociodemographic characteristics, diagnosis, referral sources, and reasons for complexity. We compared cases with and without DS (DS+ vs. DS-), and DS cases living alone and those living with family (DS+ living alone vs. DS+ living with family). We received responses from 33 IPISTs and collected data from 70 complex cases. DS was selected as the reason for complexity in 24 cases, 14 of which were people living alone. Significantly fewer DS+ cases were referred by family (8.3% vs. 54.3%), while significantly more were referred by others (87.5% vs. 41.3%). The most common referral sources were neighbours and social workers. DS+ cases had significantly higher rates of self-neglect (87.5% vs. 13.0%). Those living alone were significantly younger (median (interquartile range): 78.5 (76.8-82.8) years) than those living with family (85.0 (78.8-89.8) years). Family members living with DS+ cases also had issues, including mental illness or maltreatment. In this study sample, many older adults with DS exhibited self-neglect and had limited family support. Our findings suggest that a comprehensive approach to both older adults with DS and their families is crucial for appropriate intervention. Therefore, IPISTs that have a multidisciplinary team with an outreach function may be effective.

Objective.The generalized equivalent uniform dose (gEUD) is a well-established metric for radiotherapy dose optimization, particularly for normal tissues. However, the lack of theoretical clarity in its application has often led to empirical use in clinical practice. This study aims to reformulate gEUD-based optimization in a theoretical framework using the gEUD curve concept, and to develop a robust optimization strategy tailored for stereotactic radiation therapy (SRT), with a specific focus on bone and spinal metastases.Approach.We interpreted the gEUD as a smooth function of itsa-parameter, forming a continuous curve whose deformability decreases with increasinga. Based on this understanding, we proposed two methods: the multiple gEUD objective (MgEUDO) for optimizing the remaining volume at risk, and the selective gEUD objective (SgEUDO) for critical organs at risk. These methods were retrospectively evaluated in 22 patients who received five-fraction SRT (35 Gy total prescription).Main results.Using consistent gEUD-based optimization, all cases achieved clinically favorable dose distributions. Compared to conventional normal tissue objective constraints, the proposed strategy reduced mean dose to surrounding tissues by 10%, while improving tumor dose coverage by 0.6%. SgEUDO further achieved 39% and 37.5% mean dose reductions in the left and right kidneys, respectively.Significance.Our theoretical and practical refinement of gEUD optimization enables systematic control of dose distribution with reduced inter-planner variability. The combined MgEUDO and SgEUDO strategies provide a generalizable and clinically effective framework for high-precision radiotherapy.