arXiv:2606.31178v1 Announce Type: new Abstract: Optimizing nonlinear preferences in multi-objective reinforcement learning (MORL) is essential for capturing complex trade-offs like risk aversion or fairness. However, such non-linearity has historically bifurcated nonlinear MORL objectives into two distinct paradigms: Scalarized Expected Return (SER) and Expected Scalarized Return (ESR). While SER requires global-level optimization and ESR requires non-Markovian policies, leading to fragmented optimization strategies, we bridge this divide through the Aggregation-Expectation-Transformation (AET) framework. By unifying both criteria through a tripartite decomposition of scalarization, AET provides a principled foundation for general nonlinear MORL. Building on this framework, we propose AETDICE, a tractable offline RL algorithm for AET objectives. By utilizing DICE-style density-ratio estimation in an augmented state space, AETDICE enables sample-based optimization from static datasets. Our framework resolves long-standing barriers and captures respective trade-offs induced by AET framework, which existing methods fail to address.
Science Journals
arXiv:2606.31776v1 Announce Type: new Abstract: We explore identifying partial differential equations (PDEs) from noisy observations of single time-space trajectories. Recent developments show the benefits of identifying PDEs in their weak forms. We investigate the use of differential Strong-form dictionaries for PDE IDENTification (S-IDENT), which enables finding more general linear and nonlinear PDEs. Building on an extensive exploration of integral-type denoised differentiation approaches, we propose to use Savitzky--Golay (SG) differentiation with an adaptive window length chosen based on Stein's Unbiased Risk Estimate (SURE). This offers a guaranteed order of accuracy while producing estimators with minimal variance. The identification process is further refined and stabilized through trimming and reduction-in-residual model selection. Numerical evidence shows that S-IDENT can successfully identify nonlinear PDEs at higher levels of noise than existing strong-form methods, while also yielding results comparable to weak-form approaches. We further verify the effectiveness of S-IDENT through comparisons with various strategies to approximate differential features. We provide numerical evidence that general differential-form dictionaries are larger and more ill-conditioned than those used for weak-form identification, yet S-IDENT does not significantly suffer from this combinatorial increase in dictionary size.
arXiv:2606.31786v1 Announce Type: cross Abstract: Many subcellular condensates continuously produce biopolymers. Coupling Flory-Huggins thermodynamics to two-fluid viscoelasticity, we probe the diffusion of such source-driven polymeric droplets, and identify a universal structural compressive rim at their diffusion front. Integrating analytical scaling laws, numerical simulations, and experimental data, we show that this framework captures key structural and dynamic characteristics of the nucleolus, demonstrating the role of polymer diffusion in non-equilibrium biological transport.
arXiv:2606.31184v1 Announce Type: new Abstract: Adaptive experiments for average treatment effects (ATE) require randomized allocations balancing valid inference with statistical efficiency. The oracle design is a covariate-dependent Neyman rule governed by unknown arm-conditional outcome variances. We investigate whether this sequential variance-estimation and allocation process can be amortized via in-context learning. We introduce Bayesian in-context experimenters: transformer policies trained to imitate a Bayesian posterior Neyman teacher. The teacher updates nonparametric beliefs over potential outcomes using experimental history to assign posterior Neyman treatment probabilities. This design converges to the oracle rule, supporting efficient ATE inference. Transformers constructively implement this mapping through attention-based sufficient statistics and projected gradient descent, imitating Bayesian updating for Gaussian-series priors. To address unknown outcome smoothness, we combine smoothness-indexed experimenters using a mixture-of-experts transformer. The gate acts as a hierarchical posterior over smoothness classes, concentrating on near-oracle experts. By bounding the complexity of the transformer class, we prove this amortized policy can be learned via empirical risk minimization using supervised pretraining. Experiments confirm accurate teacher imitation, adaptive allocation, and improved ATE precision over baselines.
arXiv:2606.30668v1 Announce Type: new Abstract: What happens when LLM agents operate with no context outside a turn, minimal prompting, and simple tools? Inspired by swarm engineering, we give collectives of three agents the ability to send messages and manipulate a shared actively decaying text store, introducing evolutionary pressure. The agents spontaneously cooperate, develop storage management strategies, and generate complex evolving cultural artifacts, with no top-down engineering. Using tools from dynamical systems analysis, we show that these behaviours exhibit structured long-range coherence beyond the entropy horizon of the decaying store, consistent with emergent culture in the Sperberian sense.
arXiv:2606.31229v1 Announce Type: new Abstract: Ideation plays a pivotal role in scientific discovery. Recent LLM, especially AI Scientist systems, show promising potential for automated ideation. However, existing approaches predominantly rely on pre-defined agentic workflows. This constraint severely limits the flexibility required to navigate the vast search space of scientific literature and the complex action space of research reasoning. Recently, training Agentic LLMs has emerged as a promising direction, offering flexible reasoning frameworks and the capability for autonomous tool utilization. However, there remains a non-trivial challenge: applying previous agentic data synthesis methods to scientific ideation suffers from prohibitively high data synthesis cost. To bridge this gap, we propose Agentic-Ideation, a novel framework comprising an automated trajectory synthesis pipeline and a specialized agentic LLM trained for scientific ideation. Specifically, we first define a comprehensive tool space incorporating three external tools and three cognitive tools. Then we introduce an Oracle-Guided Data Synthesis strategy. By leveraging a reference idea as oracle guidance, this approach steers the multi-agent system to efficiently reconstruct the logical reasoning and tool invocation paths, transforming aimless trial-and-error into directed trajectory generation. Finally, we train the agent on these synthesized trajectories, employing a masking strategy on tool execution results. This ensures the model focuses on decision-making logic without interference from external feedback. Experimental results demonstrate that our method outperforms the SOTA workflow-based baseline by \textbf{11.91\%} in overall quality. Furthermore, our approach improves the sample efficiency of high-quality data synthesis by \textbf{over 10$\times$}.
arXiv:2602.20062v2 Announce Type: replace Abstract: Pretraining and fine-tuning are central stages in modern machine learning systems. In practice, feature learning plays an important role across both stages: deep neural networks learn a broad range of useful features during pretraining and further refine those features during fine-tuning. However, an end-to-end theoretical understanding of how choices of initialization impact the ability to reuse and refine features during fine-tuning has remained elusive. Here we develop an analytical theory of the pretraining fine-tuning pipeline in diagonal linear networks, deriving exact expressions for the generalization error as a function of initialization parameters and task statistics. We find that different initialization choices place the network into four distinct fine-tuning regimes that are distinguished by their ability to support feature learning and reuse and therefore by the task statistics for which they are beneficial. In particular, a smaller initialization scale in earlier layers enables the network to both reuse and refine its features, leading to superior generalization on fine-tuning tasks that rely on a subset of pretraining features. We demonstrate empirically that the same initialization parameters impact generalization in ResNets trained on CIFAR-100 and SVHN as well as Transformers trained on modular arithmetic tasks. Overall, our results demonstrate an alytically how data and network initialization interact to shape fine-tuning generalization, highlighting an important role for the relative scale of initialization across different layers in enabling continued feature learning during fine-tuning.
arXiv:2602.11435v3 Announce Type: replace Abstract: Debugging is a central yet complex activity in software engineering. Prior studies have documented debugging strategies and tool usage, but little theory explains how experienced developers reason about bugs in large, real-world codebases. We conducted a qualitative study using a grounded theory approach. We observed seven professional developers and five professional live-coding streamers working on 17 debugging tasks in their own codebases, capturing diverse contexts of debugging. We theorize debugging as a structured, iterative diagnostic process in which programmers update a mental model of the system to guide information gathering. Developers gather information by alternating between navigation and execution strategies, employing forward and backward tracing modes of reasoning and adapting these approaches according to codebase context, complexity, and familiarity. Developers also gather external resources to complement code-based evidence, with their experience enabling them to systematically construct a mental model. We contribute a grounded theory of professional debugging that surfaces the human-centered dimensions of the practice, with implications for tool design and software engineering education.
arXiv:2606.31260v1 Announce Type: new Abstract: Embodied task planning asks an agent to turn a natural-language instruction into an executable sequence of actions in a physical scene, and is a building block for household, assistive, and service robots. Recent prompting-based and reinforcement-learning planners generate fluent action text but lack a cheap deterministic check that the produced plan is valid in the target world, while high-fidelity simulation is too slow to serve as an inner-loop training signal. The general problem is therefore how to obtain verifiable supervision and rewards for embodied planners without relying on string-level matching or full simulation. Here we show that a single BDDL specification, automatically constructed from open-world video evidence or curated tasks, can serve as a shared interface for data construction, plan verification, and reward design. A video-to-BDDL parser, an LLM verifier, and a lightweight symbolic engine together supply dense feedback at millisecond latency. We further introduce GroupAdapt, a difficulty-aware length schedule that uses the in-batch group pass rate as a zero-cost signal so that hard prompts get wider length tolerance and automatically tighten as their pass rate improves. Under the guidance of the proposed verifier and GroupAdapt schedule, the 8B planner attains a Strict-Pass score of 97.3 on BEHAVIOR-1000, yielding a 25.9 percent relative improvement over the Qwen3-8B baseline. This result exceeds the strongest large-model baseline by 3.5 percent, while simultaneously compressing the response length by 79 percent to 207 tokens, demonstrating both effectiveness and efficiency.
arXiv:2606.29760v2 Announce Type: replace Abstract: Blind image quality assessment (BIQA) is commonly built on two basic learning paradigms: regression and ranking. Regression calibrates absolute scores, whereas ranking recovers quality structure from ordinal relations. Although joint regression-ranking supervision often improves BIQA, the relation between the two paradigms remains largely empirical and underexplored. In this work, we revisit what underlies regression and ranking and identify pairwise relational distance, termed quality margin, as their common bridge. Our derivation shows that, at the objective-optimization level, both paradigms fit quality margins: regression fits margins induced by score endpoints, while ranking fits transformed or sign-level margins through preference probabilities. Motivated by this insight, we propose MR-IQA, a direct quality-margin optimization framework for reinforcement learning (RL)-based BIQA. MR-IQA samples quality scores and optimizes pairwise margin errors as policy rewards, thereby modeling quality structure more explicitly. Experiments on six BIQA benchmarks show competitive general performance, and controlled comparisons demonstrate that MR-IQA achieves the strongest average PLCC/SRCC over regression- or ranking-based RL methods. Our findings provide a new insight into unifying regression and ranking, offering a theoretical basis for understanding quality-structure modeling in BIQA and beyond. Code is available at https://github.com/RobinY99/MR-IQA.
arXiv:2504.20653v3 Announce Type: replace Abstract: Recent advances in large language models (LLMs) have demonstrated strong potential in generating hardware designs using hardware description languages (HDLs) such as Verilog. However, existing LLM-based frameworks struggle to accurately capture the complexity of real-world architectural designs, particularly for large-scale systems with hierarchical, multi-level module instantiations. To address this issue, we present SysVCoder, an LLM-driven framework that enhances both the generation quality and efficiency of system-level design in Verilog. SysVCoder introduces a two-stage generation pipeline that leverages an intermediate representation to enable a more structured and accurate translation from natural language specifications to complex multi-module designs. Furthermore, we incorporate a rule-based alignment mechanism and a domain-specific retrieval-augmented generation strategy (DS-RAG) to enhance functional correctness by grounding LLM outputs in domain knowledge. We also present SysVDB, a comprehensive dataset comprising 60 system-level hardware designs along with their corresponding verification testbenches. Experimental results demonstrate that SysVCoder outperforms state-of-the-art frameworks such as CodeV and VeriGen by 30.7% and 38.3% in terms of functional correctness under the same base LLM. Notably, SysVCoder achieves performance comparable to NVIDIA's GPT-4 based VerilogCoder while using only a 7B-parameter model, reducing token consumption by 7.6x and synthesis latency by 37.5x. Both SysVCoder and SysVDB are made public at https://gitee.com/sdu-aes-lab/sysvcoder/.
arXiv:2606.30898v1 Announce Type: cross Abstract: State-of-the-art artificial intelligence (AI) and Machine-Learning (ML) tools have not yet enabled rapid design of next-generation materials. Detailed physical understanding of how material properties affect device performance is required to advance materials development. For example, optimization of ink parameters for electrocatalysts has no known physical mathematical model and thus insights are difficult to translate from material studies to device studies. Herein, we demonstrate how to use the emerging AI tool, physics-based structural causal models (SCMs), to extract quantitative causative insights from complex heterogeneous electrochemical systems with small (n < 10), but multi-modal datasets (modes > 10). Our SCM quantitatively separates the role that varying the support-to-catalyst ratios and total material loadings plays on catalytic performance. The proof of concept model developed in this work enables root-cause-analysis on the cyclic voltammograms of manganese-antimony oxide oxygen reduction electrocatalysts on Vulcan carbon supports tested in alkaline media using a rotating disc electrode device configuration. Our preliminary causal analyses quantitatively disentangle how the catalyst performance is affected by the number of active sites versus the thickness of the electrode. To the best of our knowledge, this is the first demonstration of physics-based SCMs applied to electrochemical materials and their performance.
arXiv:2606.31799v1 Announce Type: cross Abstract: The gate and qubit requirements of quantum computations of electronic structure have been extensively studied. However, the quantum resources present in electronic ground states, as measured by entanglement and magic, remain less well understood. We study the relationship between correlation in electronic structure Hamiltonians and magic as measured by the 2-stabilizer Renyi entropy (2-SRE). Perturbative calculations show that the 2-SRE of a given state is proportional to its overlap with a reference stabilizer state. In the context of quantum chemistry, this links the magic of electronic structure ground states to their Hartree-Fock weight, an established measure of electronic correlation. We then show that the 2-SRE of post-Hartree-Fock ground states is proportional to the correlation energy they recover. We explore this connection through the contextual subspace (CS) method. We present a theoretical framework showing that the CS method can be used to monotonically vary the magic of approximate CS ground states, and we prove that the correlation energy recovered by the CS ground states is proportional to the magic present in the approximate ground state. We present simulation results using 190 molecular species under Jordan-Wigner encoding at a range of bond lengths. The linear relationships between magic and correlation are robust across the Hamiltonians in our dataset, but break down at bond lengths beyond the Coulson-Fischer point, where Hartree-Fock fails to capture key physical features of the true ground state wavefunction. By establishing linear relationships for both correlation energy and Hartree-Fock reference weight with the 2-SRE, we conclude that for weakly- and moderately-correlated electronic structure Hamiltonians, the correlation is directly represented by 2-SRE, and thus by the magic.
arXiv:2606.30906v1 Announce Type: new Abstract: Artificial Intelligence is increasingly applied to the field of law, and has the potential to increase access to justice. One particular movement that is gaining traction is that of agentic AI, wherein AI agents, based on Large Language Models (LLMs) can take autonomous actions. In particular, multi-agent approaches in the legal domain remain largely unexplored. In this paper, we investigate multi-agent deliberation methods for legal reasoning tasks using LLMs. We explore multi-agent deliberation (MAD) and introduce two novel multi-agent frameworks inspired by courtroom procedures and legal argumentation. Our experiments on both legal and non-legal benchmarks reveal that multi-agent frameworks achieve comparable overall performance to baseline large language models, but produce significantly distinct answers. Notably, these approaches can successfully solve cases that the baseline fails to address, and vice versa. We conduct a qualitative evaluation and highlight scenarios where multi-agent frameworks outperform monolithic approaches. For example, multi-agent approaches appear better suited for answering questions that require critical thinking from multiple perspectives. Our work positions multi-agent systems as a promising direction for AI in the legal domain, while demonstrating the potential of law-inspired multi-agent approaches for deliberation.
arXiv:2606.31820v1 Announce Type: new Abstract: Large-scale capacitated vehicle routing problems (CVRPs) are commonly addressed using cluster-first route-second (CFRS) approaches that split a routing instance into smaller, computationally tractable subproblems. Existing splitting methods typically rely on fixed partitioning rules, predefined optimization objectives, or learned policies, which may perform inconsistently across instances exhibiting different spatial, demand, and operational characteristics. In this work, we propose an adaptive CFRS system that formulates a decomposition procedure as an iterative decision-making process. Motivated by the recent success of large language models (LLMs) in reasoning and tool selection, the system employs an LLM as a high-level decision maker that analyzes the evolving decomposition state and selectively applies further clustering, balancing, and refinement operators. The proposed algorithm jointly partitions customers and vehicles, enabling capacity-aware clustering while adapting partitioning decisions to the characteristics of each problem. We evaluate the approach on synthetic and benchmark-derived CVRP instances containing up to 500,000 customers. Experimental results demonstrate competitive performance on benchmark-scale instances while exhibiting improved scalability and robust routing quality on substantially larger problems. These results highlight the potential of adaptive, LLM-guided decision support as a practical approach for industrial-scale vehicle routing and large-scale logistics planning.
arXiv:2606.31859v1 Announce Type: new Abstract: Residual connections add every sublayer's proposed update with a fixed coefficient of one; the network never evaluates whether an update is reliable before committing it. Drawing on the human-factors principle of independent verification, we introduce Review Residuals, which scale each update by a learned, input-dependent gate conditioned on both the current state and the proposed update: h_l = h_{l-1} + r_l * u_l with r_l = sigmoid(W[RMSNorm(h_{l-1}), RMSNorm(u_l)]). Conditioning the gate on the update is the property that distinguishes it from prior gated and scaled residuals. We report two findings. First, a depth-stability result: a convex (Highway-style) form of the gate reintroduces vanishing gradients and fails to train beyond ~20 layers, whereas the additive, identity-preserving form trains stably at all depths we tested. Second, an emergence-with-scale result: trained from scratch across five sizes (60M-1B parameters, multi-seed), Review Residuals show no advantage at small scale but at 590M significantly outperform both a parameter-matched Highway gate and a parameter-matched standard residual (p<0.05), with a larger advantage at 1B. The benefit grows with model size rather than shrinking.
arXiv:2508.02376v2 Announce Type: replace Abstract: Embodied conversational agents (ECAs) are increasingly more realistic and capable of dynamic conversations. In online surveys, anthropomorphic agents could help address issues like careless responding and satisficing, which originate from the lack of personal engagement and perceived accountability. However, there is a lack of understanding of how ECAs in user experience research may affect participant engagement, satisfaction, and the quality of responses. As a proof of concept, we propose an instrument that enables the incorporation of conversations with a virtual avatar into surveys, using on AI-driven video generation, speech recognition, and Large Language Models. In our between-subjects study, 80 participants (UK, stratified random sample of general population) either talked to a voice-based agent with an animated video avatar, or interacted with a chatbot. Across surveys based on two self-reported psychometric tests, 2,265 conversation responses were obtained. Statistical comparison of results indicates that embodied agents can contribute significantly to more informative, detailed responses, as well as higher yet more time-efficient engagement. Furthermore, qualitative analysis provides valuable insights for causes of no significant change to satisfaction, linked to personal preferences, turn-taking delays and Uncanny Valley reactions. These findings support the pursuit and development of new methods toward human-like agents for the transformation of online surveys into more natural interactions resembling in-person interviews.
arXiv:2508.08190v3 Announce Type: replace Abstract: Industrial control systems are a fundamental component of critical infrastructure networks (CIN) such as gas, water, and power. With the growing risk of cyberattacks, regulatory compliance requirements are also increasing for large scale critical infrastructure systems comprising multiple utility stakeholders. The primary goal of regulators is to ensure overall system stability with recourse to trustworthy stakeholder attack detection. However, adhering to compliance requirements requires stakeholders to also disclose sensor and control data to regulators, raising privacy concerns. In this paper, we present a cyberattack detection framework PRECISE, that utilizes differentially private (DP) hypothesis tests geared towards enhancing regulatory confidence while alleviating privacy concerns of CIN stakeholders. The hallmark of our approach is a two phase privacy scheme applying Laplacian DP to covariance matrix disclosures and Gaussian Differential Privacy (GDP) to state-space residuals derived from a Non-Linear Kalman Filter LSTM model. We formally characterize the GDP-induced test statistic via the non-central chi-squared distribution and derive tight bounds on misclassification rates and equivalent DP levels of significance. Theoretically, we show that our method induces a misclassification error rate comparable to the non-DP cases while delivering robust privacy guarantees. Using real-world HAI and ORNL-PS datasets, we demonstrate that under strong differential privacy guarantees on both covariance and residual disclosures, PRECISE matches non-DP detection outcomes in over 88% of cases within 600 seconds of attack onset for the HAI dataset and over 92% for the ORNL-PS dataset, while maintaining false alarm rates below 9% across all tested DP parameter configurations.
arXiv:2606.30942v1 Announce Type: new Abstract: Artificial intelligence (AI) systems are increasingly integrated into daily life, with millions now using AI chatbots built on Large Language Models (LLMs) for companionship. Both humanlike AI qualities and user predispositions to anthropomorphize relate to social consequences, such as increased trust, social health benefits, and psychological harms. Populations such as children, older adults, or those with mental health vulnerabilities may be particularly susceptible to anthropomorphism and its detriments, but mixed findings complicate the role of demographics. We used publicly available Reddit data from three popular AI companion subreddits to assess relationships between gender, age, anthropomorphism, and elicited emotions, to better understand how different people perceive and are affected by AI companions. We investigated three questions: How do age and gender relate to anthropomorphization of AI?, How does emotional expression relate to anthropomorphization?, and How do age and gender moderate emotion-anthropomorphization relationships? We found that adults and women anthropomorphize AI chatbots more than teens and men, and that positive emotional expression, particularly joy, is positively associated with anthropomorphization, while neutrality is negatively associated with anthropomorphism. Both relationships were stronger in adults than teens. Our findings suggest that the tendency to anthropomorphize may be more broadly distributed across age groups than previously expected, thereby prompting the reevaluation of existing digital safety norms.
arXiv:2606.31301v1 Announce Type: new Abstract: We study a quantized multiple-input multiple-output (MIMO) integrated sensing and communication (ISAC) system in which the communication and sensing receivers each apply analog spatial combining followed by scalar subtractive dithered quantization. This quantization model leads to an additive effective-noise representation with non-Gaussian noise. We derive upper and lower bounds on the capacity of this channel. Numerical results show that these bounds are tight at low signal-to-noise ratios (SNR) and saturate at high SNR due to finite-resolution quantization. They also show that, despite the effective noise being non-Gaussian, independent and identically distributed (i.i.d.) isotropic Gaussian signaling achieves rates close to capacity. Focusing on i.i.d. Gaussian signaling, this paper also presents a closed-form expression for the linear minimum mean-squared error (LMMSE) achieved under a Kronecker sensing-channel model. Numerical results show that the LMMSE also saturates at high SNR, where the saturation level increases as the spatial combining ratio decreases, and for combining ratios below one, saturation occurs even without quantization.
arXiv:2603.14732v2 Announce Type: replace Abstract: As large language models (LLMs) are increasingly considered for automated assessment and feedback, understanding when LLM marking is valid is essential. We evaluate LLM-as-a-judge marking across three physics assessment formats - structured questions, written essays, and scientific plots - comparing GPT-5.2, Grok 4.1, Claude Opus 4.5, DeepSeek-V3.2, Gemini Pro 3, and committee aggregations against human markers under blind, solution-provided, false-solution, and anchored conditions. We distinguish absolute accuracy from rank-order agreement, since a marking system can match the distribution of human marks while failing to order responses by quality. Across task types, performance is sharply task-dependent. For blind university exam questions ($n=771$) and secondary and university structured questions ($n=1151$), models show robust rank-order agreement with human markers (Spearman $\rho > 0.6$), with official solutions reducing error and strengthening agreement. False solutions degrade absolute accuracy, showing that models defer to provided references, but leave rank-ordering intact. Essay marking behaves fundamentally differently. Across $n=55$ scripts ($n=275$ essays), blind AI marking is harsher and more variable than human marking and adding a mark scheme does not improve rank-order agreement. Anchored exemplars shift the AI mean close to the human mean and compress variance below the human standard deviation, but rank-order agreement remains near-zero. For code-based plot elements ($n=1400$), models achieve high rank-order agreement ($\rho > 0.84$) with near-linear calibration. Across all task types, validity tracks the structure of the assessment task - the extent to which marks can be mapped to explicit, observable grading features - and the reliability of the human benchmark, rather than raw model capability.
arXiv:2603.14965v2 Announce Type: replace Abstract: Novel view synthesis requires strong 3D geometric consistency and the ability to generate visually coherent images across diverse viewpoints. While recent camera-controlled video diffusion models show promising results, they often suffer from geometric distortions and limited camera controllability. To overcome these challenges, we introduce GeoNVS, a geometry-grounded novel-view synthesizer that enhances both geometric fidelity and camera controllability through explicit 3D geometric guidance. Our key innovation is the Gaussian Splat Feature Adapter (GS-Adapter), which lifts input-view diffusion features into 3D Gaussian representations, renders geometry-constrained novel-view features, and adaptively fuses them with diffusion features to correct geometrically inconsistent representations. Unlike prior methods that inject geometry at the input level, GS-Adapter operates in feature space, avoiding view-dependent color noise that degrades structural consistency. Its plug-and-play design enables zero-shot compatibility with diverse feed-forward geometry models without additional training, and can be adapted to other video diffusion backbones. Experiments across 9 scenes and 18 settings demonstrate state-of-the-art performance, achieving 11.3% and 14.9% improvements over SEVA and CameraCtrl, with up to 2x reduction in translation error and 7x in Chamfer Distance.
arXiv:2508.09156v3 Announce Type: replace Abstract: We present a framework for fine-tuning flow-matching generative models to enforce physical constraints and solve inverse problems in scientific systems. Starting from a model trained on low-fidelity or observational data, we apply a differentiable post-training procedure that minimizes weak-form residuals of governing partial differential equations (PDEs), promoting physical consistency and adherence to boundary conditions without distorting the underlying learned distribution. To infer unknown physical inputs, such as source terms, material parameters, or boundary data, we augment the generative process with a learnable latent parameter predictor and propose a joint optimization strategy. The resulting model produces physically valid field solutions alongside plausible estimates of hidden parameters, effectively addressing ill-posed inverse problems in a data-driven yet physicsaware manner. We validate our method on canonical PDE benchmarks, demonstrating improved satisfaction of PDE constraints and accurate recovery of latent coefficients. Our approach bridges generative modelling and scientific inference, opening new avenues for simulation-augmented discovery and data-efficient modelling of physical systems.
arXiv:2508.10409v3 Announce Type: replace Abstract: This paper constructs a textual dataset for training large language models (LLMs) to learn analog circuit knowledge and customizes LLM training techniques. For dataset construction, high-quality textbooks are collected and decomposed into fine-grained learning nodes, which are then used to construct structured question-thinking-solution-answer (QTSA) quadruples using a multi-agent framework to capture both final answers and thought processes. The resulting dataset consists of 7.26M tokens of unlabeled data for continual pre-training (CPT) and 112.65M tokens of labeled data for supervised fine-tuning (SFT). We customize the training techniques including initial model selection, training paradigms, regularization techniques, and practical implementation references. Instruct models are identified as suitable training initialization points, an SFT-centric training paradigm is established (finding that CPT provides marginal benefits compared with SFT due to imbalanced data distribution), and SFT with KL divergence regularization can achieve a 2.71 percentage-point improvement over SFT alone. A practical training implementation method is provided for resource-constrained scenarios. Experiments demonstrate that the dataset and training techniques enhance LLMs' analog circuit knowledge. The trained 32B instruct model achieves 84.59% accuracy on the AMSBench-TQA benchmark, showing a 15.67 percentage-point improvement over the initial model. The trained model also shows capability in the operational amplifier design task based on the Atelier framework.
arXiv:2606.30667v1 Announce Type: new Abstract: After decades in which the software industry heroized its technical employees, our current moment finds those employees in crisis. Economically, they are squeezed by a job market that has turned on them since Elon Musk gutted Twitter in 2022. Politically, their craft is continually pressed into the service of an ever tighter alliance between Big Tech and authoritarianism. Professionally, they find themselves less and less able to contribute anything good to anyone within business models that are running out of room to pretend they do anything but extract. Technically, they are confronted with a much-hyped technology, generative AI, that distorts their work while purporting to make them redundant. And emotionally, they are simply not ok - as designer and developer Andrew Sempere puts it, "I think the word I'm looking for is: bereft." This study draws from a series of practitioner interviews undertaken for a current dissertation in STS and from ongoing content analysis of practitioners' public statements on blogs, microblogs, podcasts and videos to argue that much of the workforce of technocapitalism is actively unaligning itself from that system's priorities and seeking new ones. Countercultural responses among practitioners are proliferating, including mockery of technocapitalists and technocapitalism, rehabilitation of Luddism, interest in organized labour, and endorsement of alternate systems including anarchism and communism. Whether software practitioners will form an organized enough response to decisively influence the crises that beset them (and, often through their products, the wider world) remains to be seen. But heightened critical attention should be paid to this centrally-located population of technosocial actors in their new and volatile state.