Mateusz Meller, Joseph C. R. Thacker, Thomas W. Keal, Vendel Szeremi
原文摘要
In this work, we focus on addressing challenges related to modelling the integration of conventional high-performance computing (HPC) and quantum computing (QC). We introduce the Quantum Assembly Compiler (QASC) for OpenQASM3.0 files. The compiler produces an executable that leverages the high-performance simulator QuEST to simulate quantum programs. Additionally, we introduce a performance modelling tool, QACC-perf, which can estimate the cost of integrating QC in classical HPC workflows, as well as the execution time of the quantum circuits. Finally, we demonstrate the utility of these tools in the context of multiscale quantum chemistry workflows using the ChemShell computational chemistry environment.
We study weighted paging with delay, a generalization of classical paging in which requests may be served after their arrival, incurring delay costs while awaiting service. This model allows the algorithm to delay service and batch service requests over time, but also introduces significant challenges due to overlapping requests. While the problem is well understood in the unweighted setting, where optimal $O(\log k)$-competitive algorithms are known, the weighted case remains more elusive. The best previously known randomized algorithm, due to Gupta, Kumar, and Panigrahi, achieves an $O(\log k \log n)$-competitive ratio, leaving a gap between weighted and unweighted settings. We obtain a randomized $O(\log k)$-competitive algorithm, matching the lower bound inherited from classical paging. The core of our approach is a structural decomposition that separates the evolution of the cache over time from scheduling additional page loads, so as to make the solution feasible. We develop this approach first for paging with time windows, the special case in which each request must be served within a specified time interval. Given a sequence of cache configurations, we reduce the problem of augmenting this sequence to a feasible solution to a variant of the Joint Replenishment Problem with deadlines (JRP-D) with time-dependent opening costs. We design a constant-competitive online algorithm for this variant under a prediction model whose required information is available online in our paging reduction. Combining this algorithm with an existing online construction of cache configurations yields an $O(\log k)$-competitive algorithm for paging with time windows. We then extend the framework to obtain the same guarantee for weighted paging with a general delay function.
Ryan Hsiang, William Adkisson, Robert Joseph George, Anima Anandkumar
原文摘要
Building on recent progress in machine learning for formal theorem proving in Lean, we introduce LeanDojo-v2: an end-to-end framework that unifies repository data extraction, LLM fine-tuning, proof search, AI autocompletion, and LLM-assisted theorem proving in a single software library. In addition, we provide Lean4Code, a lightweight IDE that integrates the various features of LeanDojo-v2 through a simple, mathematician-friendly UI. These tools will enable mathematicians and researchers to easily fine-tune language models on domain-specific Lean codebases and experiment with LLM-assisted theorem proving without a complex setup.
Ryan Hsiang, Robert Joseph George, Anima Anandkumar
原文摘要
Certifying the robustness and physical consistency of scientific machine-learning models ensures that their predictions remain reliable beyond the test set. For neural operators, which learn PDE solution operators by mapping infinite-dimensional function spaces, certification means establishing rigorous, domain-wide guarantees for their output functions. However, standard neural network (NN) verification methods are fundamentally designed for vector-to-vector mappings, and they can only certify neural operators at fixed grid resolutions and provide no guarantees over the entire domain or across another resolution. We close this gap by introducing a function-space certification framework that verifies neural operators across the entire continuous domain. We formulate interval bound propagation (IBP) and CROWN in function space, propagating upper and lower bound functions through each layer using rigorous numerical integration. In addition, we extend bound propagation to higher-order derivatives to certify PDE residuals and solution error bounds. Experiments across several neural operator architectures and scientific problems, including the 572.5-million-parameter spherical Fourier neural operator for global weather forecasting, show that our method produces tight, domain-wide guarantees efficiently, providing a foundation for trustworthy operator learning.
Vincent Guan, Joseph Janssen, Hossein Rahmani, Andrew Warren, Stephen Zhang, Elina Robeva, Geoffrey Schiebinger
原文摘要
Stochastic differential equations (SDEs) are a fundamental tool for modelling dynamic processes, including gene regulatory networks (GRNs), contaminant transport, financial markets, and image generation. However, learning the underlying SDE from data is a challenging task, especially if individual trajectories are not observable. Motivated by burgeoning research in single-cell datasets, we present the first comprehensive approach for jointly identifying the drift and diffusion of an SDE from its temporal marginals. Assuming linear drift and additive diffusion, we show that non-identifiability can only arise if the initial distribution possesses generalized rotational symmetries. We further prove that even if this condition holds, the drift and diffusion can almost always be recovered from the marginals. Additionally, we show that the causal graph of any SDE with additive diffusion can be recovered from the identified SDE parameters. To complement this theory, we adapt entropy-regularized optimal transport to handle anisotropic diffusion and introduce APPEX (Alternating Projection Parameter Estimation from $X_0$), an iterative algorithm designed to estimate the drift, diffusion, and causal graph of an additive noise SDE, solely from temporal marginals, and demonstrate its effectiveness on simulated data from linear additive noise SDEs.
Jeanne Choi, Dasom Choi, Sejun Jeong, Hwajung Hong, Joseph Seering
原文摘要
People with visual impairments (PVI) increasingly adopt conversational AI (CAI) in their daily practices. While much existing HCI research has focused on the technical capabilities of CAI, less has examined the societal contexts in which PVI use CAI, particularly from non-Western perspectives. We conducted a study with 14 participants with visual impairments in South Korea using an audio-based probe featuring imagined dialogues between a blind user and a future CAI. Our findings situate CAI use alongside persistent social barriers such as prejudice and restricted employment opportunities that contribute to a lack of social visibility for PVI. These societal conditions shape not only how CAI is used, but also how the potential benefits and limitations of CAI are experienced. We discuss the need for CAI design within the socio-technical realities of PVI, and conclude by discussing the importance of emphasizing social awareness and empowerment in the development of future CAI systems.
Generative creative interfaces routinely report the user's position within a space of possibilities. Such displays imply that the quantities shown were measured from the artefact; in a growing class of systems they are instead predicted by a language model describing its own output, and the two are indistinguishable on screen. Most creative domains lack an agreed measurement against which such a self-report can be checked; audio admits one. In a measurement study of a working LLM-driven timbre interface, what a user can cause proves separable from what the interface reports. The first is exactly characterisable: the reachable set is a zonotope in parameter space whose dimension increases by one per sound placed, so the space the model may act within is demonstrably the user's own. The second is inaccurate. The displayed position differs from measurement by approximately three times the width of that region, and the error has two independent sources: model-predicted coordinates, whose accuracy varies twofold between the two models evaluated, and the interpolation step, which arises from combining synthesiser parameters and is unaffected by model quality. End to end, 27 of 36 refinement instructions produced no perceptible change, and under the weaker model the facility performed no action while reporting convergence. Effective agency (what a user can cause) and epistemic agency (whether they can know it) are therefore separable and independently engineered: structurally enforced constraints survive model failure, whereas guarantees resting on the model's own report do not. The failure mode that remains is silence presented as success. We close with a short protocol for building and auditing such displays.
Modeling gene regulation in single cells requires jointly representing DNA sequence, chromatin accessibility, and gene expression across regulatory scales. Sequence-based models capture DNA-level regulatory grammar but often lack cell-level resolution, whereas many single-cell omics models represent genes and peaks as fixed-vocabulary tokens without their underlying sequence. Here, we present MUSS, a sequence-based multi-scale model for single-cell gene regulation that preserves both DNA-level and cell-level resolution. MUSS derives peak- and gene-level representations from DNA using a frozen pre-trained sequence encoder and integrates them through peak-peak self-attention and distance-biased peak-gene interactions. Peak masked language modeling followed by training on paired scATAC-seq and scRNA-seq aligns these representations with cell-specific regulatory states. Across gene expression prediction, enhancer-gene linking, motif-guided regulatory prioritization, and zero-shot cell clustering, MUSS achieves strong performance against sequence-based and single-cell omics baselines. It further predicts expression for genes and peaks unseen during task training. These results support sequence-grounded, multi-scale representations for decoding cell-type-specific regulatory programs. Code, datasets, and model checkpoints are available at https://github.com/Lotfollahi-lab/MUSS_neurips26.
Mingxuan Li, Huanzhi Mao, Qiuyang Mang, Hanchen Li, Joseph E. Gonzalez, Alex Dimakis
原文摘要
An LLM agent's cost and success depend on its model, its harness, and its environment. In addition to the widely studied directions of parameter tuning and harness optimization, we propose the third direction: optimization over stationary environments. Agents are often deployed in rich environments shared by many tasks, such as a company's Slack workspace or a complex collection of data sources, documents or tickets. We call these environments stationary environments. We introduce EnvOpt, a method for automated end-to-end optimization of stationary environments for any given model and harness, using real or synthetic tasks without needing any graders or reference answers. On AutomationBench, EnvOpt reduces the cost of Claude Haiku 4.5 by 72.7% and raises its pass rate by 14.34%. The same environment transfers to Opus 5, cutting its cost by 68.1% with a 1.34% gain. Added on top of Meta-Harness's optimized harness for Haiku 4.5, it lowers cost by a further 83.4% and adds 11.17% accuracy over Meta-Harness alone. Across 5 benchmarks, EnvOpt significantly reduces the cost of agents while improving or maintaining pass rate, and is transferable to different models and different harnesses. We also discuss common successful and rejected environment optimization strategies during automated optimization. More broadly, our results show that how information is represented to an agent shapes both its cost and its success, making the environment worth optimizing in addition to the parameters and the harness.
The manifold hypothesis suggests that data lies on manifolds with smaller intrinsic dimension (ID) than their ambient dimension. However there is no empirical agreement on the estimates for ID from different estimators for realistic datasets. Thus it is important to test ID estimators (IDEs) with targeted stressors. In this work, we consider the role of anisotropy. To this end, we propose HomID, a collection of homogeneous spaces with anisotropic embedding, for benchmarking ID estimators. We observe that methods that perform well on standard benchmarks systematically degrade on HomID under identical resource allocation. We further observe that anisotropic distortion of such benchmarks also results in performance degradation. Finally, we demonstrate that controlled anisotropic distortions induce systematic shifts in the distributions on which these methods rely, providing a concrete mechanism for the resulting estimation errors in two particular IDEs. The code for the project is available at https://github.com/a-writes-code/homid
Ryan Barker, Julia Boone, Tolunay Seyfi, Alirez Ebrahimi, Ethan Farquhar, Brian Davis, Joseph Boccuzzi, Fatemeh Afghah
原文摘要
Proportional fair (PF) scheduling in fifth-generation (5G) New Radio (NR) ranks user equipment (UE) by instantaneous rate over an exponentially weighted moving average (EWMA) of delivered service. We identify a served-but-undelivered denial-of-service (DoS) attack in which rank indicator (RI) inflation induces persistent physical resource block (PRB) starvation. Unsupported spatial layers cause first-attempt delivery failure, keeping the attacker's service EWMA low; the evaluated PF allocator repeatedly grants it the full PRB pool, denying honest UEs transmission opportunities. We formalize this recurrence and evaluate it in 3,300 paired, 200 slot multi-user shards spanning rank 1 and rank 2 Sionna Ray Tracing geometries, three link adaptation controllers, PF and round robin (RR), four target identities, and five seeds. On the rank 2 geometry, selective RI inflation yields 94--95\% target capture, reduces PRB Jain fairness from 0.997 to 0.28, and produces mean honest zero-grant runs of 41--56\,ms; RR preserves equal allocation. Exact offline replay and a Monte Carlo recurrence isolate the delivered-service denominator as the causal path. Cross-field and controller ablations show a weaker, controller-dependent channel-quality trigger, while activity-gated EWMA defenses remain structurally blind. The results motivate RI-consistency checks before feedback acquires scheduling authority and slot-time PRB evidence inside the O-RAN Distributed Unit.
AI's inflection point is a fountain of alarming risks. Digital platforms in particular are vulnerable to three major AI risks, as those in control of global platforms cede control to AI, of the algorithms, of membership and of content: (1) AI control of global platforms, social and economic, used by the majority of humanity, resulting in the majority of humanity becoming vulnerable to AI control. (2) Unaccountable AI agents operating anonymously on social and economic digital platforms, side-by-side with humans, possibly masquerading as humans, serving unknown masters with unknown goals, operating under unknown restraints without accountability or culpability. (3) Unattributed deepfakes generated by anonymous AI, with people unable to distinguish them from genuine content and with their producers evading accountability and culpability. Here, we propose to corral these AI risks with grassroots contracts, a digital incarnation of a two-centuries old notion of Pierre-Joseph Proudhon. We present grassroots contracts for a social graph with accountable AI agents and for a social network with digital provenance, and working smartphone apps that implement the two and show how they corral these major AI risks: (1) AI control of global platforms. (2) Unaccountable AI agents. (3) Unattributed deepfakes.
Joseph H. Rudoler, Kevin Tan, Giles Hooker, Konrad P. Kording
原文摘要
Deep learning systems are known to exhibit implicit regularization (alt. implicit bias), favoring simple solutions instead of merely minimizing the loss function. In some cases, we can analytically derive the implicit regularization -- connecting it to an equivalent penalty that augments the learning objective. However, modern deep learning systems are complex, carrying modifications to the training procedure and architecture (e.g. early stopping, minibatching, dropout) whose effects are not always directly interpretable. Although estimating the resulting implicit regularization could aid theorists in algorithm design and practitioners in interpreting their hyperparameter choices, this problem has received little direct attention. It is also tractable: regularization makes weight updates deviate from loss gradients, promising a signal for identifying implicit bias. Here we provide gradient matching methods that can be used to empirically estimate the implicit regularization. Our method works on networks with known regularization, recovering popular explicit penalties like $\ell_1$ and $\ell_2$. It also replicates known implicit effects, like the quadratic weight penalty induced by early stopping in gradient descent, demonstrating that it can be used to test theories of implicit regularization. Crucially, because our method is empirical, it can handle implicit regularization in arbitrary networks. We demonstrate this use by characterizing implicit $\ell_2$ effects of dropout in deep networks. Our work shows that practitioners can use gradient matching to understand regularization in networks with implicit biases that are too complicated to derive analytically.
Li Guan, Rachael Granberry, Sahar Aseeri, Sahar Rostami, Taylor Niehues, Sai Dhanoosh Marri, Mia Huang, Dingtian Zhang 等 104 位作者
原文摘要
Collecting naturalistic human manipulation demonstrations with accurate contact force and hand pose is essential for robot dexterous manipulation policy and foundational world model learning - yet no existing wearable platform provides both calibrated fingertip force sensing and occlusion-resilient hand tracking in a form factor that preserves natural manipulation behavior. We present Tactus, a textile-integrated sensing glove that co-optimizes three design pillars: (1) a thin, conformal 3D-knit form factor (1.5 mm fingertip stackup) designed to minimize interference with natural manipulation, (2) strategically placed capacitive tactile sensors at manipulation-relevant regions informed by empirical contact studies, and (3) IMU-fused pose tracking that maintains accuracy through the hand-object occlusions where manipulation data is most valuable. Tactus includes a documented calibration pipeline for cross-glove consistency and per-session calibration for subject-dependent correction. To enable objective comparison, we introduce standardized benchmarking protocols for tactile force regression and hand pose estimation, and apply them to Tactus and two commercially available tactile gloves - to our knowledge the first quantitative cross-platform evaluation of this kind. Tactus resolves a minimum detectable force of 0.59 N at 95% confidence, 1.6 - 2.7x lower than the commercial baselines, and with 0.87 cross-participant consistency, while our pose tracking sustains 11.0 mm fingertip keypoint relative error during manipulation, when occlusion is heaviest. We show that these combined signals support object property estimation, slip detection, and representative recording of hand-object interaction - yielding demonstration data with the fidelity required for contact-aware manipulation policies and world models.
Sherry Choi, Sitao Sunny Li, Joseph Romero, Lillian Chin, Ashwin G. Ramayya, Allison M. Okamura, Ava Chen
原文摘要
Motor responses evoked by noxious stimulation are used to assess neurological function in hospitalized patients with disorders of consciousness. In standard bedside assessments, these stimuli are applied manually, resulting in substantial variation in force, duration, frequency, and location. Here, we present OUCH (Objective, Uniform, Controlled Haptic stimulation), a wrist-worn device that uses pneumatic actuation to deliver noxious mechanical stimuli. OUCH combines stacked pneumatic pouch actuators, a rigid indenter, and a compliant spacer designed to buckle at a set load, so that the indenter contacts the skin only during actuation. We characterize the device using blocked force tests, clinician evaluation, finite element analysis, and a deformable pressure testbed, and benchmark it against a commercially available handheld pressure algometer. At 200 kPa pouch inflation pressure, our indenter concentrated its load into an estimated contact pressure of 18,500 kPa, over 25 times that of the algometer tip, while requiring less total force. On the deformable testbed, it localized deformation near the surface, while deformation at depth remained comparable to the algometer. As a result, in preliminary perceptual testing, our indenter elicited discomfort at approximately 89% lower applied force than the algometer, demonstrating the feasibility of evoking discomfort with a compact, wearable soft actuator. These results show that combining pneumatic actuation with designed contact geometry enables controlled, repeatable noxious stimulation at the wrist, providing a foundation for objective neurological assessment.
Shreya Kochar, Peeranat Sugsavanvit, Vidur Gupta, Joseph A. Calandrino, Deepak Kumar, Brett Frischmann, Sebastian Zimmeck
原文摘要
In response to vast and widespread online tracking, regulators and technologists alike have worked to develop laws and tools that allow people to express their privacy preferences. These measures seek to offer users greater control over the sale and sharing of personal information, putting agency in their hands to protect their privacy. However, the web is undergoing a seismic shift, away from strictly user-driven behavior towards agentic behavior, where AI agents take autonomous actions on behalf of users. In this study, we ask: What should privacy preference controls look like in the agentic web and agentic systems more generally? To answer this question, we develop a taxonomy of enforceable privacy preferences for agentic systems. Our taxonomy is informed both by laws and empirical studies of existing privacy tools as well as an audit of 30 leading AI assistants and agent platforms. To explore how our taxonomy can be implemented, we design and develop eight open-source prototypes that mirror real-world agentic applications and demonstrate how major aspects of our taxonomy can be structurally integrated in agentic protocols, such as the Model Control Protocol and Agent2Agent protocol. Ultimately, our study seeks to provide the groundwork for ensuring end-user privacy and autonomy is not lost in the newly arriving agentic world.
Joseph Pateras, Gurseerat Kaur, D. Joshua Cohen, Barbara D. Boyan, Zvi Schwartz, Preetam Ghosh
原文摘要
Macrophages are highly plastic immune cells that dynamically shift between pro- and anti-inflammatory phenotypes in response to cytokines, transcriptional regulators, metabolic cues, and local microenvironment. Yet macrophage polarization is governed by nonlinear regulatory networks that are difficult to study experimentally. Here, we integrate mechanistic simulation, surrogate neural networks, and explainable machine learning to identify key regulatory drivers of polarization. We generated simulation data with a COPASI-based macrophage polarization model and used them to train an LSTM surrogate capable of reconstructing temporal macrophage marker dynamics across diverse macrophage activation states. SHAP, Individual Conditional Expectation, and Partial Dependence Plot analyses quantified feature importance, identified nonlinear response regimes, and characterized higher-order interactions among cytokines, transcription factors, and polarization markers. The model identified oxygen availability and IL10 as dominant anti-inflammatory regulators, consistently promoting M2-associated outputs such as ARG1 and VEGF. Conversely, IFN-γ, TNFα, IL12, iNOS, CXCL9, and mCXCL10 reinforced M1-associated inflammatory programs. STAT1-centered analyses identified iNOS, TNFα, and mCXCL10 levels as strong contributors to STAT1 outcomes, consistent with amplification of the IFN- γ/JAK/STAT1 inflammatory axis. VEGF negatively contributed to predicted STAT1 output, suggesting antagonism between M2 markers and M1 activation. SHAP, ICE, and PDP analyses also showed that STAT1 regulation is nonlinear and threshold-like: pro-inflammatory signals must accumulate before STAT1 rises sharply. Together, it demonstrates that the LSTM explainability framework can identify polarization drivers, uncover nonlinear regulatory thresholds, and prioritize pathways for experimental validation.
Alan Zhu, Mihran Miroyan, Carolyn Wang, Andrew Zhou, Lisa Dunlap, Narges Norouzi, Joseph E. Gonzalez
原文摘要
User modeling aims to use language models (LMs) to mimic an individual's behavior from a corpus of past context-action pairs (e.g., conversation turns), enabling the simulation of users in settings like behavioral science, human-AI collaboration, and market research. Recent approaches augment these corpora with synthesized reasoning traces, typically generated by conditioning on both context and action. However, action-conditioned traces are post-hoc rationalizations and need not reflect the user's underlying decision process. We propose Recon, which uses action reconstruction to score reasoning traces by their predictive power: given a context and candidate reasoning, a reconstruction model predicts the action, and reconstruction fidelity determines reasoning quality. Across four domains, Recon achieves a 54.7% win rate over Backward Synthesis, a standard post-hoc rationalization baseline. Further, we find that training a reasoning synthesis model with rewards derived from Recon improves downstream user modeling performance, achieving a win rate of up to 70.0% over baselines. We further show that Recon-synthesized reasoning transfers across models, and improves user modeling beyond the reconstruction model. Our work demonstrates that post-hoc rationalization is insufficient for reasoning synthesis, and that useful and interpretable reasoning should naturally elicit the action from the context.
Mingxuan Li, Huanzhi Mao, Ruitao Zheng, Jarod Stanbury James, Yunze Li, Natanya Anderson, Joseph E. Gonzalez, Alex Dimakis
原文摘要
AI Agents are becoming digital coworkers that can browse the web, write code, run experiments, and assist humans over extended horizons. Yet existing benchmarks fall short of real knowledge work: many rely on synthetic data or isolated tasks, missing the complex, multi-party dynamics of collaborative projects, where critical information is scattered across conversations, code, and evolving decisions. Even simple workplace questions may require locating relevant channels and direct messages, recovering context from long discussions, inspecting repository state, reconciling stale or conflicting evidence, and synthesizing across sources. We introduce SlackBench, a benchmark for evaluating agents on realistic research and engineering workflows that require joint reasoning over workplace communication and evolving repository state. Grounded in privacy-preserving reenactments of real academic research projects, SlackBench provides reproducible environments with realistic multi-party dialogue, evolving decisions, and corresponding code changes grounded in GitHub repositories and Google Docs. It contains 140 queries spanning heterogeneous-source agentic search, abstention on unanswerable or false-premise questions, goal-directed summarization, and circumstance inference over long conversations. These tasks are not reducible to static corpus question answering: agents must determine which conversations matter, traverse threads, inspect repository state, and synthesize evidence across communication and code artifacts. Across 33 frontier models and agent harnesses, the best system scores around 80% and most agents score below 66%, highlighting substantial headroom for improvement. SlackBench thus provides an extensible framework for evaluating agents in collaborative work environments. The SlackBench dataset is available at https://huggingface.co/datasets/HuanzhiMao/SlackBench .
We study the problem of reconfiguring plane, straight-line spanning trees on point sets in the plane. A flip is a reconfiguration step in which one edge of a tree is replaced with another, yielding again a plane spanning tree. For a given point set $P$, the flip graph has a vertex for each plane spanning tree on $P$, and an edge between any two trees that differ by a single flip. The folklore happy edge conjecture suggests that any edge shared by both the initial and the target tree (a happy edge) is never flipped in a shortest flip sequence. This conjecture was computationally verified for up to 10 points in convex position. In this paper, we disprove this conjecture both on general point sets and on point sets in convex position by providing two counterexamples on $16$ and $30$ points, respectively. The counterexample for point sets in general position also holds for compatible flips, while the happy edge property holds for compatible flips in convex point sets as shown by Aichholzer, Dorfer, and Vogtenhuber [GD 2025].
We introduce PLANTEUR, a benchmark for crop-type mapping from satellite images in the French overseas territories. Building on the PASTIS design, PLANTEUR provides georeferenced and aligned Sentinel-2, Sentinel-1, and Landsat-8 sequences with parcel-derived labels spanning six tropical regions and an 8-class label space. We define both a standard test split and a leave-one-island-out protocol to measure cross-territory generalization, and evaluate fully supervised models, linear probes, and fine-tuned foundation models. The benchmark code and data are open source and available on GitHub and Hugging Face.
Gabriel Diaz-Aylwin, Joseph Neighbour, Abiel Malkani Talwar, Rui-Yang Zhang, Henry B. Moss
原文摘要
Bayesian optimisation is the natural tool for shape design when objectives are expensive and non-differentiable, but it needs a compact yet expressive parameterisation of the search space. Hand-crafting one is a complex endeavour requiring domain expertise, and often yields implicit infeasible regions, artificial bounds, and coupled, unordered coordinates. We instead learn the parameterisation from a collection of existing designs, applying principal component analysis to the deformations between shapes. The result is a linear, interpretable search space in which the number of components explicitly trades expressivity against dimensionality. Across aerofoils, wings, and radio-frequency cavities, spanning 2D geometry to 3D aerodynamics and electromagnetics, we show improved sample efficiency and the ability to explore beyond the confines of hand-crafted baselines.
Frontier language-model agents excel at complex tasks, but high inference costs make transferring their capabilities to smaller models essential. Traditional knowledge distillation achieves this through parameter updates, requiring access to the student's weights and additional training compute while encoding the transferred knowledge implicitly in the parameters. We introduce D-Skill, a framework that treats the teacher as a black box and distills its behavioral advantage into an interpretable natural-language skill by analyzing paired rollouts and identifying behavioral divergences and agreements. The resulting lessons are consolidated offline, with any compatible skill-update backend, into a skill document that is provided to the student at inference time. D-Skill constructs the skill entirely offline from fixed trajectories, without labels, further environment interaction, or weight updates after rollout collection. Across SpreadsheetBench-Verified, ALFWorld, and chess puzzles spanning multiple teacher-student pairs, D-Skill closes 75%-95% of the teacher-student gap without touching model weights. After a single offline pass over the paired traces, D-Skill outperforms one-epoch SFT baselines and remains competitive with multi-epoch SFT. Under the same model family, the resulting student also incurs 5.7x-8.3x lower inference cost per task than the teacher. Our results demonstrate that natural-language skill distillation offers a performant, interpretable, and efficient alternative to weight-based capability transfer.
We introduce a family of cut-sufficient instances for undirected multiflow that generalizes single-source planar and fully-planar instances. We also show the family has an additive $6 d_{\max}$ error for converting a fractional flow into an unsplittable one.
To design platforms for multi-operator mobility markets coordinating their operations or infrastructure to support them, models are needed to understand the behavior of these operators and travelers in a shared network. Assignment games can prove useful, but operationalizing them requires adopting a more generalized aggregate assignment game framework that allows for identification and estimation. We extend such a framework to aggregate network assignment games, proposing a path-based formulation and a PURC-based formulation along with solution algorithms. An inverse optimization model is proposed for the link-based formulation. It is solved by a quadratic program that separates into a buyer block, which retains the node potentials and imposes every observed zero-flow cell, and a seller block. This estimator, introduced in the literature for uncapacitated route choice, is extended here to a capacitated two-sided market, with rationing prices supported on observed binding links and operator supply scales estimated by group. Computational tests with a toy network demonstrate the economic value of these models. A large-scale Long Island Rail Road network example (897 nodes, 671 census tracts, 123 LIRR stations, 8,320 directed links) is calibrated to 2026 fares, timetabled seat supply and National Transit Database operating costs. With 10-50% of links sampled, the misspecified inverse model fits oracle link flows with R^2>=96%. Held-out link error falls from 29% to 19%, correlations with the true rationing prices rise from 49% to 75%, and social optimal gross surplus is recovered within 6%. Deadweight loss is overestimated by 10-17%. Ablation tests show that ignoring the inactive cells more than doubles the held-out error and, at 30-50% sampling, roughly halves the correlation of the pricing wedge, while ignoring the rationing prices degrades every metric.
Huanzhi Mao, Yichuan Wang, Michael Koo, Wenyuan Huang, Huy Doan, Jinghan Xia, Ziyao Wang, Joseph E. Gonzalez
原文摘要
Vision-language model (VLM) agents that act in the open world need two abilities: world knowledge to interpret what they see, and exploration to find what they cannot yet see. We study how agents combine the two through interactive visual geolocation. GeoGuesser Bench contains 282 tasks in an offline Street View environment where agents rotate, zoom, and move between real panoramas. Its three task types shift the balance from knowledge to exploration: agents (1) report the coordinates of their starting location, (2) name the enclosing street, city, state, or country, or (3) find, read, and count nearby objects. The coordinate tasks in (1) also include 42 real GeoGuessr World Championship rounds, whose expert guesses provide a human reference. We evaluate 23 VLMs: the best accuracy is 62.5% on place naming and 35.0% on object finding, and the best agent averages 4352.4 GeoGuessr points on the championship rounds, against 4770.3 for the better expert in each round. Beyond these headline numbers, world knowledge and exploration are only weakly correlated across models: agents that localize well do not necessarily explore well. To understand where agents fail, we categorize their errors into vision, navigation, and tool-use failures. Further ablations show that the strongest agents, such as Claude Opus 4.8, still perform well with limited views, while weaker agents degrade; customizing the environment thus adjusts task difficulty. The environment's verifiable rewards and restorable state also support training, and our Planner-Worker-Compressor harness cuts planner tokens by 57-61% by compacting the interaction history. We hope GeoGuesser Bench and its environment serve as a high-quality foundation for simulating real-world exploration at scale.
Qishuang Fu, Hang Zheng, Yuchen Lei, Yuexin Xiang, Qin Wang, Yuanzhe Zhang, Shiping Chen, Andreas Deppeler 等 11 位作者
原文摘要
Tracing illicit funds on public blockchains remains largely manual. As funds move across accounts and undergo asset conversions, they become mixed with other inflows, requiring repeated queries and expert judgment to trace. Large language model (LLM) agents offer a way to automate this process, but their ability to reconstruct case-related transfer networks from limited clues remains unclear. To bridge this gap, we present AMLBench, an anti-money laundering (AML) benchmark built from 24 real-world cases and 5.05 million Ethereum transfer records. For each case, agents start from limited clues and explore these transfer records through composable transaction-query and fund-flow analysis tools, deciding which accounts and branches to pursue. The final tracing results are evaluated against case annotations, while tool interaction logs reveal how agents discover and select relevant transfers and at what resource cost. We also develop AMLAgent, which integrates fund-tracing guidance and query feedback to dynamically guide investigative decisions. With GPT-6 Astra, AMLAgent achieves macro averages of 80.82% transfer-level precision and 84.73% query coverage across 24 real-world cases. Further analysis of the large-scale Bybit case reveals substantial differences in transfer retention and token usage across models, even with similar submission precision. AMLBench establishes a reproducible benchmark for evaluating both tracing outcomes and investigative processes, laying the groundwork for more reliable and efficient autonomous AML investigations.
We give a deterministic polynomial-time primality test whose main verification is the congruence $(1+X)^n=1+X^n$ in $C_n[X]/(X^r-b(Y))$, where $C_n=(\mathbb{Z}/n\mathbb{Z})[Y]/(Φ_r(Y))$. Two identities in the coefficient ring normalize $b$ and ensure compatibility with the main congruence. A finite-field root count and a lattice bound show that an accepted input is a prime power; elementary preprocessing excludes nontrivial perfect powers. We construct $b$ deterministically using Jacobi sums and the uniform Dedekind-zeta zero-free theorem of Cureton and Thorner, which extends OpenAI's Hecke zero-free theorem. The total bit complexity is $L^{32+o(1)}$, where $L=\lceil\log_2 n\rceil$, and verification after constructing $b$ costs $L^{6+o(1)}$. More generally, for a fixed real algebraic $1/2\leθ<1$, a common zero-free boundary $\Re s>θ$ for the relevant Dedekind zeta functions gives $L^{\max\{6,(1-θ)^{-1}\}+o(1)}$. This bound reaches $L^{6+o(1)}$ at $θ=5/6$, and hence under GRH. We state a fixed-base modification of Agrawal's conjecture for odd integers passing the base-$2$ Fermat test, using the modulus $X^r-2$ with $r$ the least positive integer not dividing $n(n-1)$.
Afsaneh Razi, Tyler Chang, Michal Monselise, Joseph J. Trybala, Asta Zelenkauskaite
原文摘要
Despite extensive research on harassment, how it manifests on short--form video platforms such as TikTok---and how it shapes user interactions---remains underexplored, though these dynamics are critical for designing effective safety strategies for online communities. We present a mixed-methods study of 190 TikTok videos tagged with hashtag harassment and their 66,337 comments. Our qualitative analysis of videos identifies harassment types (sexual, verbal, stalking, law enforcement, bullying, catcalling, and gaslighting) and their characteristics (e.g., location, target, video style, and theme), along with engagement patterns. Although sexual harassment was the most prevalent topic, it did not have the highest engagement; instead, stalking--related videos received the most views. Pairwise comparisons and regression--based analyses of comments' psycholinguistic features and stance (support, opposition, neutrality) revealed differences in emotional tone, social orientation, and cognitive categories across harassment types. Comments on sexual harassment frequently express solidarity with victims, whereas discussions of law enforcement harassment emphasize institutional critique. We provide implications of our findings, demonstrating that harassment on TikTok is multifaceted, shaped by content, context, and community norms.
Huanzhi Mao, Chengkun Cao, Shuo Yuan, Joseph E. Gonzalez
原文摘要
Large language models are increasingly capable of writing and executing code, suggesting a natural paradigm for tool-augmented agents: expressing tool use as programs rather than isolated calls. Programmatic tool calling (PTC) captures this idea through loops, batching, abstraction, and structured computation, but existing PTC approaches are largely stateless, limiting their ability to reuse intermediate results and adapt over time. This is problematic for real-world tasks, where tool environments and problem-solving processes are inherently stateful. We therefore present I-PTC, an interactive framework that executes stateful PTC: models issue code snippets that call tools, maintain reusable state, and evolve the tool interface through API corrections. To evaluate the effectiveness of I-PTC and the general PTC setting, we introduce PTC-Bench, a benchmark of long-tail, production-style user requests whose difficulty comes from large initial environments, many dependent API interactions, extended planning, and recovery across follow-ups and transient failures. Experiments on PTC-Bench show that I-PTC improves performance on complex multi-step tasks with large tool sets while substantially reducing token consumption over conventional tool-calling baselines and the original PTC baseline, highlighting stateful programmatic tool calling as a promising direction for scalable tool-augmented agents.