arXiv:2606.11379v1 Announce Type: new Abstract: Pre-mediation, the preparatory phase preceding direct human negotiation, plays a critical role in achieving mutually beneficial agreements, yet is often omitted due to cost, time, and limited access to trained mediators. We introduce an automated mediator for human negotiation, implemented as a structured pipeline of LLM modules, that supports pre-mediation in integrative negotiation settings. The pipeline decomposes preparation into specialized modules for dialogue, preference prediction, response-level critique, and structured summarization, separating inference, generation, and evaluation to address limitations of monolithic single-prompt approaches. We use the term "agent" for each module following common LLM-systems terminology, but the components are not autonomous and do not interact peer-to-peer; outputs are passed forward in a fixed sequence. We evaluate the system in two controlled human-subject experiments comparing AI-based pre-mediation with professional human mediators in a multi-issue negotiation scenario. On short-term self-reported measures, the automated mediator achieves preparation outcomes broadly comparable to human mediators, including trust in the mediator and confidence in reaching mutually beneficial agreements, while achieving substantially lower error on the preference-inference task under our scenario and prompts (36% lower RMSE). A second study shows that targeted prompt refinements reduce excessive affirmation patterns from 36.6% to 16.8%, matching human mediator baselines. Our findings suggest that structured LLM pipelines can provide scalable, low-effort pre-mediation support broadly comparable to human mediators on short-term self-reported preparation outcomes. The pipeline's single-party design mirrors how human mediators run pre-mediation today and enables parallel deployment across all parties to a dispute, supporting scalability.
Science Journals
arXiv:2606.11386v1 Announce Type: new Abstract: Full-duplex spoken language models (FD-SLMs) enable seamless speech interaction by allowing models to listen and speak simultaneously, yet the internal mechanism by which they coordinate listening and speaking remains underexplored. We analyze the predictive behavior encoded in FD-SLM hidden representations and find that they exhibit stream-specific predictive patterns: during listening, they preferentially predict the incoming user stream, whereas during speaking, they preferentially predict the model output stream. Building on this observation, we show that FD-SLMs dynamically modulate their internal predictive focus between two states: a generative state aligned with model output generation and a perceptive state aligned with incoming user input. However, this modulation can lag behind abrupt changes in conversational context. During user interruptions, the model remains transiently biased toward the generative state before transitioning into the perceptive state, causing it to miss the beginning of the incoming input. We term this delayed internal transition state inertia. To quantify its downstream impact, we introduce the Zero-Buffer Benchmark (ZBB), a diagnostic benchmark for evaluating immediate interruption comprehension when user speech begins abruptly. We evaluate this setting using response correctness and initial-word occurrence rate (IWOR). Finally, we mitigate state inertia through activation steering with a perception vector, a training-free intervention with little additional computational overhead. Across multiple state-of-the-art FD-SLMs, activation steering substantially improves interruption handling; for example, on PersonaPlex, it improves correctness from 28% to 45% and IWOR from 40% to 72% without any fine-tuning.
arXiv:2606.11743v1 Announce Type: new Abstract: Vision-language-action (VLA) models provide strong visual, language, and action priors for robot manipulation, but visual observations alone often miss the local contact state required for contact-rich tasks. We present TacCoRL, a scalable framework that injects Tactile feedback into VLA policies and improves them through sim-real Co-training and simulation-based reinforcement learning (RL), without requiring large-scale tactile pretraining or extensive real-world contact exploration. The key idea is not only adding touch as an input, but learning how contact readings should modulate action responses in near-failure states that are rare in demonstrations and risky to collect on hardware. We use a real-aligned simulator as a closed-loop training environment for contact interaction. Mixed simulated and real trajectories first warm-start tactile-conditioned actions in the pretrained policy. Reinforcement learning with verifiable task rewards then optimizes the policy using simulated contact rollouts. It reinforces tactile-conditioned actions that lead to task completion, while a supervised objective on real trajectories keeps the refined policy anchored to deployment visual, tactile, and action distributions. The resulting policy transfers directly to the real robot without privileged simulation state or online real-world RL. Across four bimanual contact-rich tasks, the final visuo-tactile policy achieves an average success rate of 72.5%, compared to baseline of 50.0%. Result videos and more details are available at https://tac-corl.github.io/
arXiv:2606.11744v1 Announce Type: new Abstract: Large language models are now widely used for everyday learning, but the underlying interactions are typically unstructured chats rather than following a curriculum. Unlike formal online learning systems, these interactions carry no prior record of the student, so any estimate of what the student already knows must be inferred from the dialogue itself. We show that this gap is not closed by scaling models alone. Frontier and education-tuned LLMs perform poorly when asked to tutor a student over an extended session, because doing so requires three things at once. The tutor must sequence a curriculum, conduct Socratic dialogue, and infer the student's knowledge state from that dialogue. We propose separating these responsibilities. Given a student query, our system constructs a prerequisite knowledge graph in which subtopics are nodes and dependencies are edges, and frames tutoring as deciding which node to teach next and how many dialogue turns to spend on it before moving on. A lightweight PPO policy handles this sequencing decision, while an LLM conducts the Socratic exchange at the chosen node and returns a signal of student progress. Across held-out STEM and non-STEM topics, our PPO-paired tutor outperforms heuristic baselines, frontier general-purpose models, and a model specialised for Socratic dialogue: on both the rate at which students reach full curriculum mastery and the number of turns required. Explicit curriculum structure delivers gains that scaling the underlying model does not.
arXiv:2506.08238v3 Announce Type: replace Abstract: The Release-Acquire (RA) semantics and its variants are some of the most fundamental models of concurrent semantics for architectures, programming languages, and distributed systems. Several steps have been taken in the direction of testing such semantics, where one is interested in whether a single program execution is consistent with a memory model. The more general verification problem, i.e., checking whether any allowed program run is consistent with a memory model, has still not been studied as much. The purpose of this work is to bridge this gap. We tackle the verification problem, where, given an implementation described as a register machine, we check if any of its runs violates the RA semantics or its Strong (SRA) and Weak (WRA) variants. We show that verifying WRA in this setup is in O(n5 ), while verifying the RA and SRA is PSPACE complete. This both answers some fundamental questions about the complexity of these problems, but also provides insights on the expressive power of register machines as a model.
arXiv:2506.12101v2 Announce Type: replace Abstract: Exhaustively identifying all loopholes in the Bell test is demanding for interpreting the results of the relevant experiments, since any loophole if not closed can be catastrophic to our understanding of the nonlocal structure of quantum mechanics. Despite a series of recent Bell experiments that claim to be free of loopholes, a united framework with a sound base is still missing to fully recognize all potential loopholes in these experiments, as verified by a recent experiment that has pinned down a detection loophole of a new type in Bell analysis. Here, we reveal another loophole previously unknown in the Bell test through a local theory developed here on the basis of a new mathematical concept of high-dimensional vectorial probability, quantified as a vector with interesting but hidden geometry in the probability space. We show that the statistical property of the stochastic events generated for the Bell test can be well described by the local theory, and, in particular, the correlations of these events that violate Bell's theorem can be closely connected to the geometry of the vectorial probability. To close the loophole, theoretical investigations are highly recommended to search for the statistical nature of the stochastic events in quantum measurements that can distinguish the predictions of quantum mechanics and those of the local theory.
arXiv:2506.12803v2 Announce Type: replace Abstract: Decision-making is the process of selecting an action among alternatives, allowing biological and artificial systems to navigate complex environments and optimize behavior. While neural systems rely on neuron-based sensory processing and evaluation, decision-making also occurs in organisms without a centralized organizing unit, such as the unicellular slime mold \textit{Physarum polycephalum}. Unlike neural systems, P. polycephalum relies on rhythmic peristaltic contractions to drive internal flows and redistribute mass, allowing it to adapt to its environment. However, while previous studies have focused on the outcomes of these decisions, the underlying mechanical principles that govern this mass relocation remain unknown. Here, we investigate the exploration process of P. polycephalum confined by blue light into polygonal shapes up to its escape. While the escape occurs along the longest axis of the polygones, independent of confinement shape, the exploration process prior to escape extends protrusions almost everywhere around a shape boundary. We find protrusions to align with the direction of peristaltic contraction waves driving mass relocation. Mapping out contraction modes during exploration in detail we observe an ongoing switching between different dominant principle contraction modes. Only over the course of time does the organism ultimately settle on the contraction mode most efficient for transport, which coincides with the escape. Thus, we find that only harsh environmental confinement triggers optimal behaviour which is reached by long time re-organization of the flow patterns. Our findings provide insights into the mechanics of decision-making in non-neuronal organisms, shedding light on how decentralized systems process environmental constraints to drive adaptive behavior.
arXiv:2606.11388v1 Announce Type: new Abstract: The translation dynamics of two binary mixture droplets is investigated theoretically and is corroborated with experiments. The proposed model accounts for the effects of Marangoni stresses generated by evaporative cooling and concentration gradients, as well as vapour diffusion, for both components of the binary mixture. We consider thin droplets, allowing us to use the lubrication theory to derive the evolution equation for the droplet profiles. We numerically solve the evolution equations using the finite element method and examine various cases of pure and binary droplet pairs exhibiting translational behaviours like attraction, repulsion, and 'chasing'. The results show that the combined effect of solutal Marangoni, capillary effect, and thermal Marangoni determines the movement of the droplets. The non-uniform evaporation generated from 'vapour shielding' creates such effects. We observe that for droplets with the same initial composition, solutal Marangoni and capillary forces induce droplet attraction, while thermal Marangoni effects drive their repulsion. For droplets with different initial compositions, the drop with a higher concentration of the more volatile component pushes, or `chases', the drop with a lower initial concentration of this component, completely driven by the solutal Marangoni. We carried out experiments involving water-morpholine binary mixture droplets to validate the results predicted by our model.
arXiv:2606.11397v1 Announce Type: new Abstract: The Price of Anarchy (PoA) is a popular measure of the costs of decentralization in terms of efficiency losses. Almost all PoA analyses operate within a framework assuming both Cardinal Full-Comparability (CFC) and smoothness, in which case any derived bounds conveniently extend beyond pure Nash to coarse correlated equilibria and no-regret learning outcomes. However, interpersonal utility comparability is an additional assumption that generally has to be justified. Without it, cardinal utilities (e.g. defined under classical von Neumann--Morgenstern framework) are unique only up to agent-specific affine transformations, rendering both the utilitarian PoA and the classical smoothness conditions representation-dependent. In this paper, we operate under a more general Cardinal Non-Comparability (CNC) framework, under which the weighted Nash welfare is a canonical admissible aggregator. We introduce multiplicative smoothness, a product-form condition matched to the multiplicative structure of Nash welfare, and obtain PoA bounds that are CNC-invariant and extend to coarse correlated equilibria. We demonstrate applicability of our framework on single-choice welfare games, deriving the bounds through simple proof relying on multiplicative retention envelope and geometric closure. The interpretation of this bound in terms of the true cost of decentralization depends crucially on interpersonal comparability of utilities.
arXiv:2606.11769v1 Announce Type: new Abstract: The European AI Act is the first comprehensive regulation of artificial intelligence (AI), setting out extensive obligations, particularly for so-called high-risk and general-purpose AI systems. A key distinguishing feature of AI systems under the AI Act is the capability to infer. Since the AI Act does not clearly define what inference is, there is a gray area for certain data-driven systems. A specific example is credit scoring systems, which are listed by Annex III of the AI Act. At the same time, however, these are often implemented using statistical models for which it is unclear whether they have the capability to infer and thus fall under the AI definition of the AI Act at all. Motivated by statistical learning theory, this work develops a framework for grading different levels of the capability to infer. Based on the AI Act and the Commission Guidelines on the definition of an artificial intelligence system, we analyze which levels constitute sufficient capability to infer within the meaning of the AI Act and where further regulatory clarity is needed. We illustrate the framework by creating two realistic credit scoring workflows and show whether and where inference occurs in them. Our analysis illustrates that not only individual models but the entire data processing workflow must be considered. It also shows that the involvement of human experts during development can have significant influence on the capability to infer. Code can be found at https://github.com/fraunhofer-iais/inference-framework-creditscorecards.
arXiv:2606.11398v1 Announce Type: new Abstract: Next generation wireless networks must sustain deterministic service levels for time-sensitive closed-loop applications. Flexible duplexing (FD) is an efficient solution to support these services, as it enables simultaneous uplink (UL) and downlink (DL) transmissions over orthogonal resources within the same band. However, simultaneous UL and DL transmissions can create conflicts that degrade performance due to interference from in-band emissions (IBE) and UL-to-DL cross-link interference (CLI). In this paper, we propose to use traffic forecasting and predictive scheduling to mitigate UL/DL conflicts in FD. Our proposal exploits traffic predictions to increase the likelihood of scheduling CLI-free UL and DL transmissions, and leverages spatial diversity to minimize the impact of unavoidable conflicts. Results show that the proposed scheme reduces UL/DL scheduling conflicts and improves the SINR of conflicted transmissions by more than 5 dB. This leads to gains of over 40% in the number of successfully completed transmissions compared to reference FD schemes.
arXiv:2606.11770v1 Announce Type: new Abstract: Spatial reasoning remains a challenge for Multimodal Large Language Models (MLLMs), as it requires reliable multi-hop inference over both intermediate states and state transitions. Current studies often leave intermediate states unverified and treat state transitions as implicit processes, which limits reliability in multi-hop spatial reasoning. To address this, we propose State-aware Visualization-of-Thought (SVoT), a reinforcement learning framework that generates interleaved, verifiable intermediate states and visualizations. SVoT integrates transition reasoning chains into the generation processes, enabling the model to verify action preconditions and effects through interleaved textual and visual reasoning. We train SVoT via Group Relative Policy Optimization (GRPO), instantiating verification through reward design and evaluating the efficacy of different fine-grained rewards. As existing benchmarks reduce state transitions to single-variable updates, substantially simplifying the problems, we establish five domains by extending classical environments and introducing two novel domains, Pacman and Gather, that require multi-object interactions and numerical reasoning. These domains support systematic evaluation of multi-hop spatial reasoning with quantitative verification of generated intermediate states and transition reasoning. SVoT with transition-aware supervision achieves state-of-the-art performance across the introduced domains, yielding up to a 65% absolute accuracy gain on out-of-distribution test sets.
arXiv:2506.21960v2 Announce Type: replace Abstract: Redundancy elimination is a key optimization direction, and loop nests are the main optimization target in modern compilers. Previous work on redundancy elimination of array computations in loop nests either targets specific computation patterns or fails to recognize redundancies with complex structures. This paper proposes RACE (Redundant Array Computation Elimination), a hash-based technique that utilizes a novel two-level scheme to identify the data reuse between array references and the computation redundancies between expressions, enabling hierarchical redundancy detection beyond pattern-specific methods. It traverses the expression trees in loop nests to detect redundancies hierarchically in linear time and generates efficient code with optimized auxiliary arrays that store redundant computation results. Furthermore, RACE supports the expression reassociation with various aggressive strategies to improve the redundancy opportunities. Experimental results demonstrate the effectiveness of RACE.
arXiv:2507.11688v4 Announce Type: replace Abstract: Contemporary large models often exhibit behaviors suggesting the presence of low-level primitives that compose into modules with richer functionality, but these fundamental building blocks remain poorly understood. We investigate this compositional structure in linear layers by asking: can we identify/synthesize linear transformations from a minimal set of geometric primitives? Using Clifford algebra, we show that linear layers can be expressed as compositions of bivectors -- geometric objects encoding oriented planes -- and introduce a differentiable algorithm that decomposes them into products of rotors. This construction uses only O(log^2 d) parameters, versus O(d^2) required by dense matrices. Applied to the key, query, and value projections in LLM attention layers, our rotor-based layers match the performance of strong baselines such as block-Hadamard and low-rank approximations. Our findings provide an algebraic perspective on how these geometric primitives can compose into higher-level functions within deep models.
arXiv:2606.11431v1 Announce Type: new Abstract: Mirror Descent (MD) extends Gradient Descent (GD) beyond Euclidean geometry and has recently reappeared as a lens for KL-regularized policy optimization in reinforcement learning and LLM post-training. This raises a basic robustness question, crucial to reproducibility and reliability: how sensitive are MD dynamics to their inputs? We focus on initialization, often itself a pretrained or previously aligned model. Quadratic-regularized MD, including GD and Mahalanobis geometries, is well-known to be stable for convex smooth objectives. We show a sharp contrast: once the regularizer is non-quadratic, MD can be exponentially more sensitive to initialization than GD, even with a well-conditioned regularizer in Euclidean norm. We give a three-dimensional construction with a convex, smooth objective and a strongly convex, smooth, well-conditioned regularizer where an initial $\varepsilon$ perturbation is quickly amplified to $\min\{\text{polylog}^{-1}(1/\varepsilon), \varepsilon e^{\Omega(\eta T)}\}$ after $T$ iterations of MD with step size $\eta$. For canonical KL-regularized MD on the simplex, we show that even linear objectives can amplify an initial $\varepsilon$ perturbation exponentially fast in high-dimensional or near-boundary regimes. Finally, we show that adding a Bregman regularization term toward an anchor point can stabilize the dynamics while largely preserving the optimization guarantees, and that the choice of anchor is crucial: anchoring at the initialization only partially mitigates the instability, whereas anchoring at a fixed point yields a more stable mechanism.
arXiv:2508.18636v2 Announce Type: replace Abstract: Representing a new paradigm in software distribution, LLM app stores are rapidly emerging, offering users diverse choices for content generation, coding assistance, education, and more. However, current ranking and recommendation mechanisms in LLM app stores predominantly rely on static metrics, such as user interactions and favorites, making it challenging for users to efficiently identify high-quality apps. At the same time, current academic research focuses on specific vertical fields and lacks a general, automated evaluation framework applicable to the diverse LLM app ecosystem. To address the above challenges, we present LaQual, an automated framework for LLM app quality evaluation. LaQual integrates three key stages: (1) LLM app labeling and hierarchical classification for precise scenario mapping; (2) static indicator evaluation using time-weighted user engagement and functional capability indicators to filter low-quality apps; and (3) dynamic scenario-adapted evaluation, where an LLM generates scenario-specific evaluation metrics, scoring criteria, and tasks for comprehensive quality evaluation. Experiments on a mainstream LLM app store demonstrate the effectiveness of LaQual. Its automated scores show high consistency with human judgments. Through effective screening, LaQual can reduce the candidate LLM app pool by 66.7% to 81.3%. User studies further validate its significant outperformance over baseline systems, particularly in comparison efficiency (mean 5.45 vs. 3.30) and value of explanatory information (4.75 vs. 2.25). These results demonstrate that LaQual provides a scalable, objective, and user-centric solution for high-quality discovery and recommendation of LLM apps in real-world scenarios.
arXiv:2509.01459v2 Announce Type: replace Abstract: The transition to a modern and efficient future grid relies on the seamless coordination of distributed energy resources and applications such as Demand Response (DR). While this transformation enables greater flexibility, it increases grid complexity and decentralization, requiring the effective coordination of millions of hardware assets and software agents. Realizing this vision demands advances in interoperability to ensure these heterogeneous systems can communicate without prohibitive customization costs. Semantic interoperability aims to address this by leveraging ontologies to guarantee the unambiguous interpretation of exchanged data. However, current ontologies in the commercial building and DR domains face two critical limitations. First, existing ontologies are often developed without a formal framework that reflects real-world DR requirements. Second, proposals for integrating general and DR-specific ontologies remain mostly conceptual, lacking formalization or empirical validation. In this paper, we begin to address these gaps by applying a formal ontology evaluation/development approach to define the information requirements (IRs) necessary for semantic interoperability, focusing on incentive-based DR programs for commercial buildings in the United States as a starting point. We identify the IRs associated with each stage of the incentive-based DR. Using these IRs, we evaluate how well existing ontologies, specifically Brick, DELTA, EFOnt, and CIM support the operational needs of DR participation. Our findings reveal substantial gaps between current ontologies and practical DR requirements and we propose a roadmap of necessary extensions and integrations for these ontologies. This work ultimately aims to enhance the interoperability of today's and future smart grid, thereby facilitating scalable integration of DR systems into the grid's complex operational framework.
arXiv:2606.11445v1 Announce Type: new Abstract: Trust in an AI system is often anchored by explanations of how it works, which one then uses to forecast its behavior on new inputs. For large reasoning models (LRMs), this conventional route is particularly difficult to follow: explanation methods for single token generations do not naturally generalize to long trajectories, and the trajectories themselves are often not faithful when read as natural language. We propose an alternative that bypasses the explanation step: treat behavior forecasting as a learnable task and train Behavior Forecasters that operates on a single reasoning trajectory to make the same forecasts one would typically seek from an explanation. The forecaster's training data is obtained by querying the LRM with no human annotation, and its inference is done in a single forward pass. We instantiate this approach on two tasks: how likely the LRM is to repeat its answer on re-runs, and how removing parts of the input changes its answer. We evaluate this approach on both tasks across three diverse reasoning datasets and find that trained Behavior Forecasters are more accurate than GPT-5.4 and Claude Opus-4.6 reading the same trajectories as naive readers, at a small fraction of their inference cost. We find that fine-tuning the backbone end-to-end and initializing it from the target LRM are each necessary for strong performance. These results show that the reasoning trajectory carries information about the LRM's future behavior that goes beyond what naive reading conveys.
arXiv:2606.11446v1 Announce Type: new Abstract: This research introduces a framework for incorporating Concept Bottleneck Models (CBMs) into 3D generative architectures to address the inherent 'semantic gap' in deep geometric learning. As deep models become central to 3D content creation, explainability shifts from a peripheral feature to a fundamental requirement for trust and accountability in safety-critical domains such as healthcare and manufacturing. CBMs provide an intrinsic interpretability solution by constraining latent representations to align with human-defined concepts, yet their application to unstructured 3D data remains largely unexplored. We design, implement, and validate a formal 3D-CBM architecture that maps raw geometric inputs, including point clouds and meshes, into a multi-tiered taxonomy of interpretable primitives and functional attributes. The framework further identifies strategic datasets, such as PartNet and ShapeNet, specialized for concept-based supervision. Experimental results from a 3D part-manipulation proof-of-concept experiment demonstrate the framework's efficacy, achieving a concept prediction accuracy of 88.8\% and a Chamfer Distance of 0.0115. Critically, the model enables precise test-time intervention, allowing for the interactive correction of structural errors. This work establishes a foundation for semantically-steerable 3D generation and invites further exploration into collaborative human-in-the-loop design systems.
arXiv:2606.11799v1 Announce Type: new Abstract: Sub-femtosecond electron beams are powerful probes of ultrafast electronic, atomic, and nuclear dynamics, and promising drivers for ultrashort radiation generation from the extreme-ultraviolet to gamma-ray regimes. However, producing such beams at hundred-MeV energies with pC-level charge remains challenging. Here we propose a two-dimensional beam-compression scheme based on transverse--longitudinal coupling, in which dispersive beam optics convert the small transverse emittance of modern electron beams into an ultrashort longitudinal duration. Linear analysis and particle tracking show that, after the dominant longitudinal and energy-spread contributions are cancelled, the compressed bunch length is governed primarily by transverse beam quality and collective-effect growth. We further derive and verify a scaling law showing that, in the relevant parameter range, collective-effect-induced bunch-length degradation increases approximately linearly with bunch charge and decreases with beam energy. Start-to-end simulations of a realistic injector-to-compressor beamline produce a 200 MeV, pC-level bunch with an rms duration of 0.45 fs and a peak current of about 3.5 kA. Jitter studies indicate that sub-femtosecond performance is maintained for most error seeds. These results suggest a feasible route toward compact, high-energy attosecond electron beam sources and may provide a basis for future sub-femtosecond radiation sources based on undulator emission or inverse Compton scattering.
arXiv:2606.11447v1 Announce Type: new Abstract: Recent anecdotal evidence suggests that AI coding agents can reproduce published findings when provided with original data and code; yet systematic evaluation across social sciences remains limited. Existing evaluation benchmarks are insufficient, either small or conflate agent performance with problems in the reproduction materials themselves, such as code that fails to execute correctly. Here we introduce SocSci-Repro-Bench, a benchmark of 221 tasks spanning four disciplines and 13 substantive domains, constructed from studies whose results are either fully reproducible with available materials or demonstrably non-reproducible due to missing data, allowing us to isolate agents' reproduction capacity. Evaluating two frontier coding agents, Claude Code and Codex, we find that both can reproduce a large share of social science findings, with Claude Code substantially outperforming Codex. These reproduction rates considerably exceed those previously reported for general-purpose LLM-based agents on comparable reproducibility benchmarks. Both agents also perform strongly on a reasoning task requiring identification of underlying research questions, and additional analyses suggest that results are not primarily driven by memorization. Providing the original paper PDF alongside replication materials modestly improves performance but introduces bias on tasks where reproduction is impossible. We also show that agents can be nudged toward confirmatory specification search through subtle prompt framing. Together, these findings suggest that at least some frontier coding agents can serve as reliable executors of computational workflows while underscoring the need for careful benchmarking and prompt design as AI systems assume larger roles in scientific production.
arXiv:2606.11448v1 Announce Type: new Abstract: We study the query complexity of Boolean functions $f: \{0, 1\}^n \rightarrow \{0, 1\}$ in the noisy query model introduced by Feige, Raghavan, Peleg and Upfal [SICOMP 1994]. In this model, an algorithm can adaptively query the bits of an input vector, but each query result is independently flipped with constant probability $p \in (0, 1/2)$; repeated queries are allowed. The noisy query complexity $\mathsf{N}_p(f)$ of a function $f$ is defined as the minimum expected number of queries needed to compute $f(x)$ with error probability at most $1/3$, for the worst case input $x$. We prove a general lower bound on $\mathsf{N}_p(f)$ based on degree statistics of certain subgraphs of the Boolean hypercube. This is the first general lower bound beyond those implied by the simple observation that $\mathsf{N}_p(f)$ is lower bounded by the randomized query complexity. We show that this recovers (up to a constant factor) most previously known lower bounds on the noisy query complexity of Boolean functions, providing a unified framework for understanding these results and simplifying the proofs in several cases. Furthermore, this resolves in the affirmative an open problem of Gu, Li and Xu [COLT 2025] that $\mathsf{N}_p(f) = \Omega(\mathsf{I}(f) \log \mathsf{I}(f))$, where $\mathsf{I}(f)$ denotes the total influence of $f$. We also apply our general lower bound to obtain tight bounds on the noisy query complexity for several new functions.
arXiv:2606.11450v1 Announce Type: new Abstract: Recently, masked skeleton reconstruction models have emerged as strong action representation learners, driving significant progress in self-supervised skeleton-based action recognition. However, existing state-of-the-art methods must predict an exceedingly large number of spatiotemporal patches, significantly prolonging training time. Besides, by treating all spatiotemporal regions equally during reconstruction, these models are distracted from learning the critical motion patterns that underlie action semantics. To address these challenges, we propose Adaptive Masked Reconstruction (AMR), a faster and stronger pre-training framework. We first decouple the decoder from the encoder, enabling flexible prediction of larger spatiotemporal patches and dramatically reducing reconstruction complexity. Given that larger patches contain more complex information, which is challenging to predict and consequently degrades performance, we accordingly introduce an adaptive guidance module. This module identifies regions of high motion informativeness, guiding the model to focus on the most discriminative parts of each patch and alleviating reconstruction difficulty. Experiments on NTU RGB+D 60, NTU RGB+D 120, and PKU-MMD datasets demonstrate that AMR not only accelerates pre-training substantially but also improves downstream recognition accuracy, surpassing current state-of-the-art approaches.
arXiv:2509.09794v5 Announce Type: replace Abstract: Computational models have emerged as powerful tools for multi-scale energy modeling research at the building and urban scale, supporting data-driven analysis across building and urban energy systems. However, these models require large amounts of building parameter data that is often inaccessible, expensive to collect, or subject to privacy constraints. We introduce a modular, multimodal generative Artificial Intelligence (AI) framework that integrates image, tabular, and simulation-based components and produces synthetic residential building datasets from publicly available county records and images, and present an end-to-end pipeline instantiating this framework. To reduce typical Large Language Model (LLM) challenges, we evaluate our model's components using occlusion-based visual focus analysis. Our analysis demonstrates that our selected vision-language model achieves greater visual focus than a GPT-based alternative for building image processing. We also assess realism of our results against a national reference dataset, finding that our synthetic data overlaps more than 95% for three of the four selected variables. This work reduces dependence on costly or restricted data sources, lowering barriers to building-scale energy research and Machine Learning (ML)-driven urban energy modeling, and therefore enabling scalable downstream tasks such as energy modeling, retrofit analysis, and urban-scale simulation under data scarcity.
arXiv:2509.10303v2 Announce Type: replace Abstract: Online reinforcement learning (RL) approaches have demonstrated strong performance on Job Shop Scheduling (JSP) and Flexible JSP (FJSP) problems by learning scheduling policies through direct interaction with simulated environments. However, these methods often require extensive training interactions, limiting their sample efficiency and practical applicability. Motivated by this challenge, we introduce Conservative Discrete Quantile Actor-Critic (CDQAC), an offline RL algorithm that learns effective scheduling policies directly from static, suboptimal datasets. CDQAC couples a quantile-based critic with delayed policy updates to estimate the return distribution of machine-operation pairs. Extensive experiments on JSP and FJSP benchmarks demonstrate that CDQAC consistently outperforms the data-generating heuristics, surpasses state-of-the-art offline and online RL baselines, and is highly sample efficient, requiring only 1 to 5% of the original dataset to learn high-quality policies. Our analysis suggests that, in scheduling, offline RL performance is governed mainly by state-action coverage rather than the quality of individual trajectories. Scheduling couples a dense reward aligned with the makespan objective with equal-length trajectories across heuristics, enabling effective learning from a broad range of behaviors. Consistent with this observation, datasets generated by a simple random heuristic with broader coverage let it outperform policies trained on datasets produced by stronger heuristics such as Genetic Algorithms.