Category
442 papers
The designs produced by Mise-en-Sc\`ene are the closest to the ground truth in perceived quality among all compared methods, by a wide margin over both an LLM layout planner and a specialized layout transformer, while the match-and-place stage bridges the remaining fidelity gap to the ground-truth composites.
Small- and medium-sized enterprises (SMEs) increasingly need to reconfigure how they create, deliver, and capture value in response to digitalisation, sustainability demands, and environmental uncertainty. This study systematically reviews 162 English-language articles published in SSCI-indexed journals to integrate fragmented research on business model innovation (BMI) in SMEs. Descriptive analysis, keyword co-occurrence analysis, and article-level qualitative content analysis are combined. The findings identify four broad knowledge domains and show that 41 studies treat BMI primarily as an outcome, 44 as an organisational process, 17 as an antecedent of subsequent outcomes, and 36 as part of a combined causal relationship; a further 24 studies are descriptive or typological. Across these roles, the literature explains BMI through resource mobilisation, entrepreneurial action logics, absorptive and dynamic capabilities, experimentation, and network mobilisation. Its effects on performance, growth, resilience, internationalisation, and sustainability depend on complementary resources, implementation capabilities, stakeholder alignment, and environmental conditions. The evidence supports an adaptive and iterative interpretation of SME BMI, but direct evidence of recursive feedback remains limited. Resource constraints also have conditional effects, stimulating bricolage and focused experimentation in some circumstances while inhibiting substantive change when shortages become severe. The review provides an integrative synthesis specific to SMEs and advances digital business model innovation research by explaining how digital pressures are translated into changes in value proposition, value creation and delivery, and value capture through organisational and relational mechanisms.
This survey presents a critical review of VLMs for egocentric video understanding, tracing the progression from conventional recognition architectures to multimodal foundation models and embodied systems, and examines how first-person perception and multimodal foundation models support wearable assistance, robot skill learning, human-to-robot transfer, and embodied decision making.
Mechanistic interpretability seeks quantities that models do not expose directly: represented states, component effects, interactions, and responses to interventions. Patching, gradients, Hessian-vector products, and subset interventions provide different measurements under different access assumptions and may target different quantities. We formulate their shared measurement structure as mechanistic tomography: designed measurement for recovering internal mechanisms and intervention effects. For a chosen basis and intervention family, measurements take the form y = Ax + w, where A describes the interventions, x is the target map, and w contains nonlinear response, sampling error, and basis misspecification. This language gives a practical procedure: start with the least costly measurements, test on held-out interventions at the intended scale, calibrate simple mismatch, and expand the measurement family when structured residuals remain. Control provides a demanding validation setting because an estimate that guides an intervention acts as an observer. In a two-HMM model, control error rises with observer error, while target improvement can hide nuisance-state movement. Under forward-only access, sparse aggregate measurements recover a finite-effect map with fewer interventions than coordinate patching. With gradient access, finite probes improve a local attribution map. Lifted measurements and Hessian-vector products recover interactions missed by first-order maps, while Tracr shows that the required family depends on the basis. On GPT-2-small IOI, the Name Mover-Negative Name Mover interaction is the largest held-out predictive term among three tested cross-group pairs. On Qwen-2.5-7B, finite calibration makes an additive refusal-response map adequate, so held-out error does not support pairwise lifting.
Reach audiences
Advertise in front of researchers, engineers, and readers.
The results show that referent selection and boundary precision are partially separable, with different components moving opposing regions of the IoU curve -- behavior a single threshold cannot reveal.
A conventional all-attention model of the same size on the same data and a conventional all-attention hybrid that beats GPT-2 124M, Pythia-160M, OPT-125M and GPT-neo-125M, and exceeds MobileLLM-125M's published score despite that model seeing a trillion tokens.
This project built a pipeline that pulls news articles from the News API, company background from Wikipedia, and stock price data from Yahoo Finance for ten major companies and developed a simple but effective template that converts stock data into natural language narratives.
G3Ego, a graph-based framework for egocentric action understanding that uses gaze as a structural cue to identify action-relevant entities in the scene, achieves competitive performance compared with video-based approaches and consistently improves Macro-F1 under class-imbalanced evaluation, while avoiding reliance on computationally expensive video pretraining.
SafeBranch is proposed, a framework that aligns an embodied actor on safety through branch pairs constructed from the actor's own unsafe rollouts via environment rollback, achieving roughly ten times more safe successes than the untrained baseline on the unseen-object variant.
Overall, the study shows how conversational surveys, structured data processing, conventional behavioral modeling, machine learning, and multimodal LLM prediction can be coordinated within an auditable multi-agent workflow.
Auditing three rounds of rank-$32$ LoRA self-training on Qwen3-8B against a frozen control pushed through the identical pipeline, this work identifies seven measurement failures, each of which inverts a reported finding when its control is absent.
A rapidly advancing precision-therapy pipeline-including antisense oligonucleotides to upregulate the intact allele, AAV-based gene replacement, CRISPR-mediated transcriptional activation, epigenetic modulators, and rational pathway-targeted small molecules-offers realistic prospects for disease modification.
Data Management
What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration. The post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Earth AI
Health & Bioscience
