Science communication revision requires writers to dynamically balance scientific exposition and narrative engagement - a process where writers often struggle with competing directions. Existing LLM-assisted tools help with co-writing, but offer limited support for navigating this iterative, multi-directional revision process. To address this gap, we designed Spatial Balancing, an exploratory revision environment that maps rhetorical goals and revision strategies onto a two-dimensional spatial canvas for experienced science communication creators with domain expertise but lacking formal professional training. By building a design space of communication strategies and embedding them into a spatial exploratory canvas, our system treats feedback as navigational cues rather than prescriptive judgments. Our findings show that this integrated revision environment helps writers stay focused on writing goals, reason about revision as trajectories, and explore alternatives, which supports greater metacognitive control and confidence without increasing workload. This work highlights the value of spatially externalized revision environments for supporting iterative, reflective thinking during LLM-assisted writing.

Designing cycling infrastructure requires balancing the competing needs of diverse user groups, yet designers often struggle to anticipate how different cyclists experience the same street environment. We investigate how persona-based evaluation can support cycling infrastructure design by making experiential conflicts explicit during the design process. Informed by a formative study with 12 domain experts and crowdsourced bikeability assessments from 427 cyclists, we present StreetDesignAI, an interactive system that enables designers to (1) ground evaluation in real street context through imagery and map data, (2) receive parallel feedback from simulated cyclist personas spanning confident to cautious users, and (3) iteratively modify designs while the system surfaces conflicts across perspectives. A within-subjects study with 26 transportation professionals comparing StreetDesignAI against a general-purpose AI chatbot demonstrates that structured multi-perspective feedback significantly Broaden designers’ understanding of various cyclists’ perspectives, ability to identify diverse persona needs, and confidence in translating those needs into design decisions. Participants also reported significantly higher overall satisfaction and stronger intention to use the system in professional practice. Qualitative findings further illuminate how explicit conflict surfacing transforms design exploration from single-perspective optimization toward deliberate trade-off reasoning. We discuss implications for AI-assisted tools that scaffold persona-aware design through disagreement as an interaction primitive.

Most AI systems today are designed to manage tasks and execute predefined steps. This makes them effective for process coordination but limited in their ability to engage in joint problem-solving with humans or contribute new ideas. We introduce MultiColleagues, a multi-agent conversational system that shows how AI agents can act as colleagues by conversing with each other, sharing new ideas, and actively involving users in collaborative ideation processes. In a within-subjects study with 20 participants, we compared MultiColleagues to a single-agent baseline. Results show that MultiColleagues fostered stronger perceived social presence, and participants rated their outcomes as higher in quality and novelty, with more elaboration during ideation. These findings demonstrate the potential of AI agents to move beyond process partners toward colleagues that share intent, strengthen group dynamics, and collaborate with humans to advance ideas.

High-Level Synthesis (HLS) provides a fast path from concepts to silicon, but practical HLS design flows begin with a tedious step: refactoring software into HLS-compatible programs. Converting real-world software remains challenging due to restrictive language support and the gap between software and hardware programming practices, and this preparatory phase can take domain experts days even for an initial synthesizable design. Existing automated refactoring methods and recent LLM-based workflows partially address this problem, yet they often fall short in generalizability, scalability, and cost efficiency.

High-level synthesis (HLS) is a widely used tool in designing Field Programmable Gate Array (FPGA). HLS enables FPGA design with software programming languages by compiling the source code into an FPGA circuit. The source code includes a program (called ``kernel'') and several pragmas that instruct hardware synthesis, such as parallelization, pipeline, etc. While it is relatively easy for software developers to design the program, it heavily relies on hardware knowledge to design the pragmas, posing a big challenge for software developers. Recently, different machine learning algorithms, such as GNNs, have been proposed to automate the pragma design via performance prediction. However, when applying the trained model on new kernels, the significant domain shift often leads to unsatisfactory performance. We propose a more domain-generalizable model structure: a two-level hierarchical Mixture of Experts (MoE), that can be flexibly adapted to any GNN model. Different expert networks can learn to deal with different regions in the representation space, and they can utilize similar patterns between the old kernels and new kernels. In the low-level MoE, we apply MoE on three natural granularities of a program: node, basic block, and graph. The high-level MoE learns to aggregate the three granularities for the final decision. To stably train the hierarchical MoE, we further propose a two-stage training method. Extensive experiments verify the effectiveness of the hierarchical MoE.

Figure 1: Our system supports nonlinear AI-assisted hypothesis exploration that balances breadth and depth of exploration.Using the node-link diagram shared representation and integrated information hint panels that display preliminary results and related work, a participant in our user study deeply explored a branch of hypotheses around gender income gaps, including a nuanced hypothesis about field-specific gender disparities in income (A); but also simultaneously kept track of the overall hypothesis space, and backtracked to explore other hypothesis branches around variations in income by marital status (B), such as post-divorce impacts on income.A more detailed version Figure 10 can be found in Appendix B.

Unsustainable behaviors are challenging to prevent due to their long-term, often unclear consequences. Serious games offer a promising solution by creating artificial environments where players can immediately experience the outcomes of their actions. To explore this potential, we developed EcoEcho, a GenAI-powered game leveraging multimodal agents to raise sustainability awareness. These agents engage players in natural conversations, prompting them to take in-game actions that lead to visible environmental impacts. We evaluated EcoEcho using a mixed-methods approach with 23 participants. Results show a significant increase in intended sustainable behaviors post-game, although attitudes towards sustainability had only marginal effects, suggesting that in-game actions likely can motivate intended real world behaviors despite similar opinions on sustainability. This finding highlights multimodal agents and action-consequence mechanics to effectively raising sustainability awareness and the potential of motivating real-world behavioral change. © 2025 Copyright held by the owner/author(s).

With the continuous development of generative AI’s logical reasoning abilities, AI’s growing code-generation potential poses challenges for both technical and creative professionals. But how can these advances be directed toward empowering junior researchers and designers who often require additional help to build and express their professional and personal identities? We introduce Frontend Diffusion, a multiagent coding system transforming user-drawn layouts and textual prompts into refined website code, thereby supporting selfrepresentation goals. A user study with 13 junior researchers and designers shows AI as a human capability enhancer rather than a replacement, and highlights the importance of bidirectional human-AI alignment. We then discuss future work such as leveraging AI for career development and fostering bidirectional human-AI alignment of multi-agent systems.

Recently, memory consumption of Deep Neural Network (DNN) rapidly increases, mainly due to long lifetimes and large shapes of tensors. Graph scheduling has emerged as an effective memory optimization technique, which determines the optimal execution, re-computation, swap-out, and swap-in timings for each operator/tensor. However, it often hurts performance significantly and can only manipulate tensors' lifetimes but not shapes, limiting the optimization space. We find that graph transformation, which can change the tensor shapes and graph structure, creates a new trade-off space between memory and performance. Nevertheless, graph transformation are applied separately so far, with primary focus on optimizing performance and not memory.

Interactive systems that facilitate exposure to examples can augment problem solving performance. However designers of such systems are often faced with many practical design decisions about how users will interact with examples, with little clear theoretical guidance. To understand how example interaction design choices affect whether/how people benefit from examples, we conducted an experiment where 182 participants worked on a controlled analog to an exploratory creativity task, with access to examples of varying diversity and presentation interfaces. Task performance was worse when examples were presented in a list, compared to contextualized in the exploration space or shown in a dropdown list. Example lists were associated with more fixation, whereas contextualized examples were associated with using examples to formulate a model of the problem space to guide exploration. We discuss implications of these results for a theoretical framework that maps design choices to fundamental psychological mechanisms of creative inspiration from examples.

There has been a growing trend in deploying deep neural networks (DNNs) on tiny devices. However, deploying DNNs on such devices poses significant challenges due to the contradiction between DNNs' substantial memory requirements and the stringent memory constraints of tiny devices. Some prior works incur large latency overhead to save memory and target only simple CNNs, while others employ coarse-grained scheduling for complicated networks, leading to limited memory footprint reduction. This paper proposes MoteNN that performs fine-grained scheduling via operator partitioning on arbitrary DNNs to dramatically reduce peak memory usage with little latency overhead. MoteNN presents a graph representation named Axis Connecting Graph (ACG) to perform operator partition at graph-level efficiently. MoteNN further proposes an algorithm that finds the partition and schedule guided by memory bottlenecks. We evaluate MoteNN using various popular networks and show that MoteNN achieves up to 80% of peak memory usage reduction compared to the state-of-art works with nearly no latency overhead on tiny devices.

There have been several recent works proposed to utilize model-based optimization methods to improve the productivity of using high-level synthesis (HLS) to design domain-specific architectures. They would replace the time-consuming performance estimation or simulation of design with a proxy model, and automatically insert pragmas to guide hardware optimizations. In this work, we address the challenges associated with high-level synthesis (HLS) design space exploration (DSE) through the evolving landscape of HLS tools. As these tools develop, the quality of results (QoR) from synthesis can vary significantly, complicating the maintenance of optimal design strategies across different toolchains. We introduce Active-CEM, a task transfer learning scheme that leverages a model-based explorer designed to adapt efficiently to changes in toolchains. This approach optimizes sample efficiency by identifying high-quality design configurations under a new toolchain without requiring extensive re-evaluation. We further refine our methodology by incorporating toolchain-invariant modeling. This allows us to predict QoR changes more accurately despite shifts in the black-box implementation of the toolchains. Experiment results on the HLSyn benchmark transitioning to new toolchain show an average performance improvement of 2.38× compared to AutoDSE and a 1.2× improvement over HARP, while also increasing the sample efficiency by 5.75×, and reducing the runtime by 2.7×.

The ongoing trend of hardware specialization has led to a growing use of custom data formats when processing sparse workloads, which are typically memory-bound. These formats facilitate optimized software/hardware implementations by utilizing sparsity pattern- or target-aware data structures and layouts to enhance memory access latency and bandwidth utilization. However, existing sparse tensor programming models and compilers offer little or no support for productively customizing the sparse formats. Additionally, because these frameworks represent formats using a limited set of per-dimension attributes, they lack the flexibility to accommodate numerous new variations of custom sparse data structures and layouts. To overcome this deficiency, we propose UniSparse, an intermediate language that provides a unified abstraction for representing and customizing sparse formats. Unlike the existing attribute-based frameworks, UniSparse decouples the logical representation of the sparse tensor (i.e., the data structure) from its low-level memory layout, enabling the customization of both. As a result, a rich set of format customizations can be succinctly expressed in a small set of well-defined query, mutation, and layout primitives. We also develop a compiler leveraging the MLIR infrastructure, which supports adaptive customization of formats, and automatic code generation of format conversion and compute operations for heterogeneous architectures. We demonstrate the efficacy of our approach through experiments running commonly-used sparse linear algebra operations with specialized formats on multiple different hardware targets, including an Intel CPU, an NVIDIA GPU, an AMD Xilinx FPGA, and a simulated processing-in-memory (PIM) device.

Cross-domain analogical reasoning is a core creative ability that can be challenging for humans. Recent work has shown some proofs-of-concept of Large language Models’ (LLMs) ability to generate cross-domain analogies. However, the reliability and potential usefulness of this capacity for augmenting human creative work has received little systematic exploration. In this paper, we systematically explore LLMs capacity to augment cross-domain analogical reasoning. Across three studies, we found: 1) LLM-generated cross-domain analogies were frequently judged as helpful in the context of a problem reformulation task (median 4 out of 5 helpfulness rating), and frequently (∼ 80% of cases) led to observable changes in problem formulations, and 2) there was an upper bound of ∼ 25% of outputs being rated as potentially harmful, with a majority due to potentially upsetting content, rather than biased or toxic content. These results demonstrate the potential utility — and risks — of LLMs for augmenting cross-domain analogical creativity.

Online whiteboards are becoming a popular way to facilitate collaborative design work, providing a free-form environment to curate ideas. However, as templates are increasingly being used to scaffold contributions from non-experts designers, it is crucial to understand their impact on the creative process. In this paper, we present the results from a study with 114 students in a large introductory design course. Our results confirm prior findings that templates benefit students by providing a starting point, a shared process, and the ability to access their own work from previous steps. While prior research has criticized templates for being too rigid, we discovered that using templates within a free-form environment resulted in visual patterns of free-form curation where concepts were spatially organized, clustered, color-coded, and connected using arrows and lines. We introduce the concept of ‘Free-form Templates’ to illustrate how templates and free-form curation can be synergistic.

Advanced packaging offers a new design paradigm in the post-Moore era, where many small chiplets can be assembled into a large system. Based on heterogeneous integration, a chiplet-based accelerator can be highly specialized for a specific workload, demonstrating extreme efficiency and cost reduction. To fully leverage this potential, it is critical to explore both the architectural design space for individual chiplets and different integration options to assemble these chiplets, which have yet to be fully exploited by existing proposals. This paper proposes Monad, a cost-aware specialization approach for chiplet-based spatial accelerators that explores the tradeoffs between PPA and fabrication costs. To evaluate a specialized system, we introduce a modeling framework considering the non-uniformity in dataflow, pipelining, and communications when executing multiple tensor workloads on different chiplets. We propose to combine the architecture and integration design space by uniformly encoding the design aspects for both spaces and exploring them with a systematic ML-based approach. The experiments demonstrate that Monad can achieve an average of 16% and 30% EDP reduction compared with the state-of-the-art chiplet-based accelerators, Simba and NN-Baton, respectively.

Conversational agents (CAs) have the great potential in mitigating the clinicians’ burden in screening for neurocognitive disorders among older adults. It is important, therefore, to develop CAs that can be engaging, to elicit conversational speech input from older adult participants for supporting assessment of cognitive abilities. As an initial step, this paper presents research in developing the backchanneling ability in CAs in the form of a verbal response to engage the speaker. We analyzed 246 conversations of cognitive assessments between older adults and human assessors, and derived the categories of reactive backchannels (e.g. “hmm”) and proactive backchannels (e.g. “please keep going”). This is used in the development of TalkTive, a CA which can predict both timing and form of backchanneling during cognitive assessments. The study then invited 36 older adult participants to evaluate the backchanneling feature. Results show that proactive backchanneling is more appreciated by participants than reactive backchanneling.

Problem framing—the process of defining a problem—has been described by many researchers and designers as the crux of the design process. However, novice designers struggle with problem framing. To better understand this process and the potential for scaffolding, we conducted two studies. In the first study, we analyzed 41 problem statements from an introductory design course and found that novices often omit key information, like the primary stakeholder or the obstacles they face. To get novices to reflect on and include necessary design information, we created a tool called ProbLib that cues novices to explicitly reflect on aspects of the problem such as the stakeholders. To evaluate this approach, we conducted a between-subjects study (N=73) to compare ProbLib with an unstructured open text form. We found that participants using ProbLib wrote higher quality statements, included more information, and were more confident about specifying design needs. We observed creative behaviors such as brainstorming and analogical reasoning.

When exploring a new domain through web search, people often struggle to articulate queries because they lack domain-specific language and well-defined informational goals. Perhaps search tools rely too much on the query to understand what a searcher wants. Towards expanding this contextual understanding of a user during exploratory search, we introduce a novel system, CoNotate, which offers query suggestions based on analyzing the searcher’s notes and previous searches for patterns and gaps in information. To evaluate this approach, we conducted a within-subjects study where participants (n=38) conducted exploratory searches using a baseline system (standard web search) and the CoNotate system. The CoNotate approach helped searchers issue significantly more queries, and discover more terminology than standard web search. This work demonstrates how search can leverage user-generated content to help people get started when exploring complex, multi-faceted information spaces.

While research shows that web search plays a role throughout the creative process, less is known about about how people use web search to learn and frame their thinking about an open problem. People need web search to gather information about a problem area, but this can also influence the rest of the creative process. To understand how web search affects early-stage design, we collected and analyzed search log and self-report data from 34 students in a project-based design class. Participants reported struggling with scoping broad, ill-defined information goals into queries, learning domain-specific language, and assessing the usefulness of information. Analysis found that more active and diverse search behavior (i.e. issuing more frequent and diverse queries, and opening more webpages) related to more progress in early-stage design (i.e. gathering more facts, articulating more insights, and developing better problem frames). Based on these findings, we discuss implications for designing search tools to support peoples' creative processes.

Multi-label classification remains a challenging problem due to the inherent label imbalance issue, which brings overfitting of minor categories to modern deep models. In this paper, to tackle this issue, we propose a novel method named Variational Feature Augmentation (VFA) to enhance the deep neural networks for multi-label classification. Our method decouples the feature vectors extracted by the backbone network into multiple low-dimensional spaces via a novely proposed Variational Feature Decoupling Module. The decoupled feature vectors are then re-combined with a shuffle operation and a Feature Augmentation Layer to enrich the minor co-occurrence relations, mitigating the label imbalance. Different from most other methods, VFA does not modify the network architecture or introduce extra computation cost in inference phase. We conduct comprehensive experiments on four benchmarks of two visual multi-label classification tasks, pedestrian attribute recognition and multi-label image recognition, and the results demonstrate the effectiveness and generality of the proposed VFA.

Many organizations have adopted design processes that integrate community voices to discover the real problems that communities face. Online discussion forums offer a familiar and flexible technology that can help facilitate discussion around problems and potential solutions. However, we lack understanding about what information community members share, how that information is structured, and how social interactions affect design processes at scale. This paper presents a mixed-methods analysis of Canvas, a learning management system, which enables users to contribute to the design of the platform by sharing and deliberating on problems and solutions in a discussion forum. We collected and analyzed 1412 ideas and 18,335 associated comments shared on the Canvas discussion forum. We found that the distributed nature of design information, the presence of duplicate ideas, and contributors' gaming behaviors made it difficult for the community to make sense of the design discussion. These gaming behaviors also constitute a new concern for participatory design research. Finally, we reflect on how Canvas community members contribute information to a shared design space and how future systems could more effectively coordinate community design efforts.

Due to the significant variability in waveforms and characteristics of ECG signals, developing fully automatic (i.e., requires no expert assistance) heartbeat classification algorithms with satisfactory performance on domain-shifted data remains challenging. In this letter, we propose a novel Mixup Asymmetric Tri-training (MIAT) method to improve the generalization ability of heartbeat classifiers in domain shift scenarios. First, we develop an ECG-based tri-branch CNN model, including one shared feature encoder followed by three branch networks. Next, to obtain target-discriminative features progressively, the tri-branch CNN is trained asymmetrically in each domain adaptation cycle, where two branches are used to assign pseudo-labels to the target domain samples and the third branch is trained on these pseudo-labeled target samples. Moreover, three kinds of mixup regularizations are incorporated into the training process. Experimental results on MITDB and SVDB show that the proposed MIAT outperforms the state-of-the-art methods in terms of F1-macro score and demonstrate the effectiveness of each mixup regularization.

Electrocardiogram (ECG) delineation, which provides clinically useful information for the diagnosis of cardiovascular disease, is an essential task in automated ECG analysis. The discrepancies among ECG signals from different datasets, namely domain shifts, may bring severe challenges to the cross-dataset performance of ECG delineation algorithms. The domain shifts are generally caused by the differences of conditions, collecting devices, and individual characteristics, and are inherent and non-negligible in ECG. In this work, we propose an unsupervised domain adaptation method called Adaptive Region Aggregation Network (ARAN) based on adversarial training to tackle domain shift problem in ECG delineation. The proposed algorithm promotes the state- of-the-art deep neural network RAN[1] to learn domain- invariant features and achieve improving performance on both source and target domain. The experiments results on two public datasets, LUDB and QT database, prove that our approach can effectively improve the cross-dataset performance of the state-of-the-art deep learning model.

The alteration of DNA methylation landscape is a key epigenetic event in cancer. As the accumulation of large-scale genome-wide DNA methylation data from clinical samples, we are able to characterize the patterns of DNA methylation alterations for identifying candidate epigenetic markers and drivers. In this survey, we take hepatocellular carcinoma (HCC) as an example to show the basic steps of analyzing the DNA methylation patterns in cancer across multiple data sets. We collected three genome-wide DNA methylation data sets with ∼800 clinical samples and the corresponding gene expression data sets. First, by quantitatively analyzing two global methylation alterations, it is found that about 90% tumors acquire either genome-wide DNA hypo-methylation or CpG island methylator phenotype. Second, probe-level analysis identified 267, 228 and 197 hyper-methylated sites in promoter regions for the three data sets, respectively. These local hyper-methylated patterns are highly consistent: 84 sites (from 61 promoters) are hyper-methylated in all the three studied data sets, including many previously reported genes, such as CDKL2, TBX15 and NKX6-2. Then, these hyper-methylated sites were used as candidate markers to classify tumor and non-tumor samples. The classifiers based on only 10 selected probes can achieve high discriminative ability across different data sets. Finally, by integrative analyzing DNA methylation and gene expression data, we identified 222 candidate epigenetic drivers, which are enriched in inflammatory response and multiple metabolic pathways. A set of high-confidence candidates, including SFN, SPP1 and TKT, are significantly associated with patients' overall survivals. In summary, this study systematically characterized the DNA methylation alterations and their impacts on gene expressions in HCCs based on multiple data sets.

MOTIVATION: Molecule-based prediction of drug response is one major task of precision oncology. Recently, large-scale cancer genomic studies, such as The Cancer Genome Atlas (TCGA), provide the opportunity to evaluate the predictive utility of molecular data for clinical drug responses in multiple cancer types. RESULTS: Here, we first curated the drug treatment information from TCGA. Four chemotherapeutic drugs had more than 180 clinical response records. Then, we developed a computational framework to evaluate the molecule based predictions of clinical responses of the four drugs and to identify the corresponding molecular signatures. Results show that mRNA or miRNA expressions can predict drug responses significantly better than random classifiers in specific cancer types. A few signature genes are involved in drug response related pathways, such as DDB1 in DNA repair pathway and DLL4 in Notch signaling pathway. Finally, we applied the framework to predict responses across multiple cancer types and found that the prediction performances get improved for cisplatin based on miRNA expressions. Integrative analysis of clinical drug response data and molecular data offers opportunities for discovering predictive markers in cancer. This study provides a starting point to objectively evaluate the molecule-based predictions of clinical drug responses. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: MicroRNAs (miRNAs), a class of small regulatory RNAs, play important roles in cancer initiation, progression and therapy. MiRNAs are found to regulate diverse cancer-related processes by targeting a large set of oncogenic and tumor-suppressive genes. To establish a high-confidence reference resource for studying the miRNA-regulated target genes and cellular processes in cancer, we manually curated 2259 entries of cancer-related miRNA regulations with direct experimental evidence from ∼9000 abstracts, covering more than 300 miRNAs and 829 target genes across 25 cancer tissues. A web-based portal named oncomiRDB, which provides both graphical and text-based interfaces, was developed for easily browsing and searching all the annotations. It should be a useful resource for both the computational analysis and experimental study on miRNA regulatory networks and functions in cancer. AVAILABILITY AND IMPLEMENTATION: http://bioinfo.au.tsinghua.edu.cn/oncomirdb/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.