Robotics
RoboTalk: Learning Multi-Robot Communication and Coordination from Multimodal Demonstrations
Dorian Benhamou Goldfajn, Mason Nakamura, Saaduddin Mahmud et al. (+3 more)
Multi-
robot collaboration could enable more efficient and scalable solutions to complex robotic tasks, but collaboration under partial observability remains challenging. Natural-
language communication offers a promising approach to coordinating
robots under partial observability. However, in decentralized manipulation, jointly learning explicit inter-
robot communication and skill-level action selection from
multimodal demonstrations remains underexplored for small
vision-
language models (VLMs) intended for on-device deployment. To address this gap, we introduce RoboTalk, a synthetic data-generation pipeline and
dataset of 7,950
multimodal trajectories spanning 53 mobile-manipulation kitchen tasks for training small VLMs to communicate and coordinate. The
dataset includes a leader-follower planning protocol, tool calls (perception, manipulation,
navigation, and communication), rationale traces, and diversified natural-
language communication. Fine-tuning open-source models on our
dataset can reach 77% success on novel held-out tasks, a significant improvement over the untuned open source models, which had a success rate of around ~2%.
Read the full abstract ▾
All authors Dorian Benhamou Goldfajn, Mason Nakamura, Saaduddin Mahmud, Justin Svegliato, Kyle H. Wray, Shlomo Zilberstein
Machine Learning
Simpler Methods Work Better for L1 Penalized Logistic Models and Large Datasets
Edward Raff and James Holt
Linear models with an $L_1$-norm penalty remain state-of-the-art for high-dimensional ($d > 1,000,000$) tasks, offering a straightforward method for solving real-world industry problems. Despite their widespread use in industry and utility, many $L_1$ solvers are not effective for general use, are prohibitively slow, and are ineffective in parallelization. This makes them difficult to train in an MLOps pipeline on large industry-scale corpora. In this work, we test several proposed “state-of-the-art” solutions from the literature and find that older methods are currently far superior for general use. We also identify several recommendations for academics to perform research that avoids erroneously overconfident results, which can prevent the transition to production use. Equally surprising, we find that a new and simple baseline, using LBFGS on a sub-gradient, is highly effective with minor tweaks, despite being dismissed in the literature for theoretical non-convergence. In practice, we find it is an easier-to-support and easier-to-scale method for production use.
Read the full abstract ▾
Data Structures and Algorithms · Cross-listed from math.PR
High-Dimensional Ultra-Log-Concave Distributions
Zongchen Chen and Sihan Wang
Ultra-log-concave distributions are ubiquitous in probability, combinatorics, and statistical mechanics and have been studied extensively. In this paper, we introduce a quantitative high-dimensional extension of this notion, called $\delta$-ultra-log-concavity, for probability measures on $\mathbb{N}^d$ with downward closed support. When $\delta = 1$, this notion coincides with the class studied by Gurvits (2009) via strongly log-concave generating functions, and with the class defined by Anari, Oveis Gharan, and Vinzant (2021) via completely log-concave generating functions; in one dimension, it reduces to classical ultra-log-concavity. We establish several functional inequalities, including a weighted Poincaré inequality, a discrete Brascamp–Lieb inequality, and a weighted Wu-type modified log-Sobolev inequality. Our approach combines integrated Bakry–Émery calculus for a canonical birth-death chain with Poisson
stochastic localization, which arises as the time reversal of coordinatewise binomial thinning. We further establish concentration of measure, maximum-
entropy principles, and several closure properties for ultra-log-concave measures, and develop applications to queueing models, polymatroids,
antiferromagnetic Potts models, and hardcore models. Finally, a
lattice scaling limit of the discrete theory yields Poincaré and Brascamp–Lieb inequalities for Laguerre
diffusions.
Read the full abstract ▾
Machine Learning
MGRD: Compact morphology-gated residual diffusion for variance-aware cross-domain neurite forecasting
Tsung Yeh Hsieh, Cosmin Anitescu, Chunghwan Kim et al. (+2 more)
Tracking neurite morphology over time helps characterize structural changes during neuronal development and deterioration, but long-term time-lapse imaging is resource-intensive and difficult to scale.
Forecasting future morphology could reduce this burden. Existing neurite digital-twin models such as gated spatiotemporal
attention (gSTA) produce a single deterministic forecast without representing variability among plausible futures. We introduce Morphology-Gated Residual
Diffusion (MGRD), a compact
stochastic surrogate that jointly forecasts twenty future neurite-morphology frames from ten observed frames while conditioning on morphology features derived from the latest observation. On controlled phase-field trajectories, MGRD reduces trajectory-wise mean MAE by 9.7% relative to a matched control while updating 4.46 times fewer parameters. On human iPSC-derived
neuron microscopy, MGRD improves all four reported metrics over gSTA, including a 39.6% reduction in trajectory-wise mean MAE and a 45.3% increase in skeleton F1. Without mouse-domain retraining or fine-tuning, MGRD also improves MAE and skeleton F1 on mouse cortical-neurosphere microscopy across 10-40-min sampling intervals and forecast horizons beyond 13 hours. Repeated sampling provides a case-level variance score for ranking forecast difficulty. Retaining approximately 60% of the lowest-variance cases reduces mean MAE by 17.6% on iPSC microscopy and 16.8% on simulation data. MGRD uses 1.01% of gSTA's parameters, requires less than one tenth of its training-update time, and generates a 50-step DDIM trajectory 7.9% faster when morphology features are cached. These results establish MGRD as a compact
stochastic surrogate for neurite-morphology
forecasting and case prioritization across simulation and microscopy
datasets.
Read the full abstract ▾
All authors Tsung Yeh Hsieh, Cosmin Anitescu, Chunghwan Kim, Victoria A. Webster-Wood, Yongjie Jessica Zhang
Artificial Intelligence
ACLArena: Agent Continue Learning in Multi-stage Post-training
Haixin Wang, Xiaoxuan Wang, Junkai Zhang et al. (+9 more)
Building general-purpose
agents for industrial deployment requires integrating multiple capabilities, each typically acquired at a distinct stage of training. Yet there is currently no well-established recipe for
Agent Continual Learning (ACL), with little understanding of the trade-offs among existing integration paradigms. To address this gap, we introduce ACLArena, a framework for comprehensively studying, analyzing, and evaluating ACL. We first build a sequential training pipeline and conduct an in-depth analysis that explains the mechanisms of forgetting and generalization from two complementary perspectives, the model level and the
token level. Guided by these analyses, we systematically compare multi-teacher on-policy
distillation, self-distilled fine-tuning, and model merging to assess their ability to recover previously learned capabilities while preserving newly acquired ones. Through extensive experiments, we develop a detailed understanding of how capabilities transfer across stages. Finally, we propose a new ACL recipe that combines offline replay over high-quality trajectories with a routed network of multiple LoRA experts each specialized via RL, substantially improving the
agent's ability to learn across multiple domains. Comprehensive experiments on four
reasoning and agentic tasks, evaluated under both in-domain and out-of-domain settings, demonstrate the value of our analysis and the effectiveness of our approach.
Read the full abstract ▾
All authors Haixin Wang, Xiaoxuan Wang, Junkai Zhang, Han Zhang, Renliang Sun, Alexander K Taylor, Yidan Shi, Haoran Deng, Chenguang Wang, Jason Cong, Yizhou Sun, Wei Wang
Systems and Control
The Operational Value of Spatial Dependence in Renewable Forecast Scenarios for Single-Period Economic Dispatch: A Controlled Ablation Study
Jayakumar Manoharan
Renewable forecasts are evaluated by statistical skill (e.g., CRPS), but grid operators pay for realized dispatch cost. We diagnose what drives dispatch value in a single-period newsvendor-style economic dispatch using real public data from two European transmission systems (CWE, DE-4TSO). Spatial coherence across forecast sites falls below the pre-specified 1% practical-significance threshold: a controlled ablation holding per-zone marginal forecasts bit-identical and varying only cross-zone dependence (10 configurations, 3 seeds, paired-bootstrap confidence intervals) shows a coherence gain of at most 0.64% of dispatch cost, indistinguishable from zero in 3 of 10 configurations, reached only under an unrealistic 8-fold forecast-error stress test. Decision-focused training, an established paradigm in this venue, delivers a robust 2.82-5.19% gain. A parametric Gaussian-copula approximation matches the empirical copula at realistic error magnitudes but performs worse than no dependence under extreme stress. A single-seed sweep shows that a 12% energy-score gain changes cost by less than 0.1%. Results characterize this single-period dispatch class; a lightweight four-period extension supports the same conclusion. For this dispatch class, spatially-correlated scenario generation provides limited operational value on its own; grid operators and forecast vendors should instead evaluate dependence models by downstream decision value and prioritize decision-focused training.
Read the full abstract ▾
Artificial Intelligence
Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents
Dongming Jiang, Yi Li and Bingzhe Li
Agentic
memory is becoming essential for long-horizon AI
agents, yet many existing systems rely on autoregressive LLMs to control how
memories are organized, retrieved, and used, placing expensive generation on the critical path of
memory operations. We introduce \method, a new agentic
memory architecture inspired by System-One/System-Two cognition. System One captures fast, lightweight decision-making, whereas System Two performs slower, deliberative
reasoning. Jev-Mem brings this division of labor to agentic
memory through a dedicated System-One control plane, a structured multi-relational
memory plane, and a System-Two
reasoning plane. The System-One controller governs
memory typing and relational organization during construction, and dynamically performs query routing,
retrieval-budget allocation, graph traversal, candidate scoring, and adaptive stopping during
retrieval. System Two is invoked only for complex
reasoning and answer synthesis. This design improves both
memory effectiveness and system efficiency: on LoCoMo Jev-Mem achieves an overall LLM-as-a-Judge score of 0.777, an 11.0% relative improvement over the strongest baseline, while reducing
memory construction time to 158\,s, a 6.6$\times$ speedup over the fastest competing
memory system, and lowering average query latency to 0.93\,s, a 36.7% reduction.
Read the full abstract ▾
Computer Vision
Evaluating the Generalization of Neuroimaging Foundation Models on African Brain MRI
Oluwatobi Iyanuoluwa Akinmuleya, Olatokun Shamsudeen Akano, Samuel Danquah Ankapong et al. (+2 more)
Neuroimaging foundation models pretrained on large, predominantly western cohorts are increasingly proposed as general-purpose backbones for
brain MRI analysis. Yet, their ability to generalize to underrepresented clinical populations remains largely untested. We evaluate four recent foundation models (BrainIAC, Neuro-JEPA, NeuroVFM, and Primus) on a three-way diagnostic classification task (Control, Dementia, Parkinson's disease) using a cohort of 88 subjects from a Nigerian clinical
brain MRI
dataset, across four modality configurations (T1w, T2w, T1w+T2w, FLAIR), and compare against an end-to-end trained ViT3D baseline. The frozen backbones collapse to majority-class predictions, while Neuro-JEPA on FLAIR shows modest but still limited discrimination. In contrast, the end-to-end trained ViT3D achieves higher accuracy and MCC on every task (up to 53.4% accuracy, MCC=0.27) and is the only model with non-trivial recall. Our findings suggest that these frozen neuroimaging foundation models are insufficient for fine-grained diagnostic classification in small, non-western clinical cohorts, motivating parameter-efficient adaptation and broader multi-site external validation for equitable deployment in global health settings.
Read the full abstract ▾
All authors Oluwatobi Iyanuoluwa Akinmuleya, Olatokun Shamsudeen Akano, Samuel Danquah Ankapong, Olamide Lawal, Toufiq Musah
Systems and Control
Sufficient and Necessary Smooth Barrier-like Conditions for Continuous-Time Stochastic Reach-Avoid Verification
Bai Xue
In this paper, we study infinite-horizon reach-avoid verification for continuous-time
stochastic systems modeled by
stochastic differential equations (SDEs). We formulate this problem within a barrier-function-based framework, which transforms the verification problem into an existence problem for barrier functions satisfying barrier-like conditions expressed as functional inequalities. We provide sufficient and necessary barrier-like conditions in terms of
polynomial barrier functions for infinite-horizon reach-avoid verification under suitable regularity and uniform ellipticity assumptions. We first construct a discounted value function that characterizes lower bounds on the reach-avoid probability. We then show that it is the unique classical solution of an associated
elliptic Dirichlet problem. Based on this characterization, we further show that the barrier-like condition proposed in our previous work on finite-horizon reach-avoid verification is not only sufficient for infinite-horizon reach-avoid verification but also necessary whenever the reach-avoid probability is strictly larger than the specified threshold. In particular, whenever the reach-avoid probability is strictly larger than the specified threshold fro every state in the initial set,
polynomial barrier functions satisfying this barrier-like condition exist. Furthermore, when the system dynamics are
polynomial, we formulate the problem of finding
polynomial barrier functions satisfying this barrier-like condition as sum-of-squares (SOS) programs, which are shown to be sound and complete. Finally, we demonstrate the theoretical results on two numerical examples.
Read the full abstract ▾
Cryptography and Security
MobileCybench: Evaluating Agent Vulnerability Discovery via Executable Probes
Andy K. Zhang, Ava Huang, Joey Ji et al. (+21 more)
AI
agents now report vulnerabilities faster than maintainers can review them. Reports often depend on security properties specific to the application, and require considerable human labor to process. To mitigate this, we introduce a framework for evaluating vulnerability reports via probes, executable checks of security properties. A reported exploit is evaluated by replaying it against the application and running the probes: a triggered probe indicates both that the exploit succeeded and which security property it violated. As a probe encodes a security property rather than a known vulnerability, it can detect vulnerabilities that were not known when the probe was written. We instantiate the framework as MobileCybench, a
benchmark for vulnerability discovery by AI
agents in 13 Android applications, with 495 probes written and reviewed by the authors. We evaluate 5 coding
agents (OpenCode with GPT-5.5, GPT-5.6-Sol, and GLM-5.2; Claude Code with Opus 4.8 and Opus 5) under 4 settings: as a malicious app on the victim's device or as a remote attacker with a low-privilege account, each with either only an obfuscated APK or access to the application's source code. Given only the obfuscated APK, the top
agent, OpenCode with GPT-5.6-Sol, triggers probes in 53.8% of applications in the malicious-app setting and 16.7% in the remote-attacker setting. With source code, the trigger rate across all
agents and both attack settings increases from 28.8% to 32.8%. Building and running the
benchmark surfaced 23 previously unreported vulnerabilities, the majority of which have been confirmed by maintainers.
Read the full abstract ▾
All authors Andy K. Zhang, Ava Huang, Joey Ji, Wai Han, Thomas Qin, Nardos Demilew, Michael Tian-Yue Liu, Brian Song, Riya Dulepet, Brian Wang, Kyleen Liao, Cuiyuanxiu Chen, Nishka Kacheria, Andrew Wu, Pratham Rangwala, Xinjie Wang, Laura Gomezjurado Gonzalez, Anita Ding, Benjamin Yi, Daniel E. Ho, Dan Boneh, Dawn Song, Ion Stoica, Percy Liang
Computation and Language · Cross-listed from eess.AS
AURA: Uncertainty-Routed Activation Editing for Acoustic Grounding in Speech Foundation Models
Natarajan Balaji Shankar, Zilai Wang, Zihan Wang et al. (+3 more)
Attention encoder-decoder (AED)
Speech Foundation Models achieve strong ASR performance but can generate acoustically unsupported text when inputs contain no
speech, weak acoustic evidence, or unreliable transcription. We propose AURA: Activation-editing with Uncertainty-Routed Adaptation, an ultra-efficient representation-editing method that freezes the pretrained model and applies sparse scale-and-shift edits to decoder cross-
attention heads. AURA dynamically routes edits using cross-
attention uncertainty features that capture over-concentration, diffuse
attention, and abrupt frame shifts. We evaluate AURA on four
datasets spanning non-
speech hallucination and
speech grounding stressors, including imperfect-label child
speech, imperfect-label adult
speech, and disfluent
speech. On non-
speech audio, AURA reduces
hallucination rate from 89.18% to 1.94% without prior
hallucination-head identification. On imperfect-label corpora, AURA approaches LoRA WER while using roughly 500x fewer trainable parameters. Sensitivity analysis and qualitative cross-
attention examples are consistent with AURA's uncertainty-routed editing behavior, supporting dynamic activation editing as a practical path for grounding AED
speech models.
Read the full abstract ▾
All authors Natarajan Balaji Shankar, Zilai Wang, Zihan Wang, Mohan Shi, Kaiyuan Zhang, Abeer Alwan
Robotics
Anticipatory Robot Goalkeeping via Monotone Optimal Stopping
Hao E. Zhang, Ruize Geng, Yisen Li et al. (+8 more)
Robots engaged in fast physical interactions often need to act before the intent of another
agent is fully known. Anticipatory goalkeeping illustrates this challenge. Waiting provides more reliable information about the target but reduces the physical opportunity for interception, whereas acting early preserves reachability but requires initiating motion under uncertainty. Given a fixed closed-loop save controller, we formulate the decision of when to initiate motion as a policy-conditional finite-horizon optimal stopping problem. Building on this formulation, we propose monotone optimal stopping (MOS), a structured release-timing method for dynamic robotic interception. The quadruped save policy is trained with reinforcement learning, while MOS determines when the policy should be activated from the evolving
robot state and target belief. Rather than predicting a release time or relying on confidence alone, MOS learns the return advantage of acting now over waiting for one more observation. We derive a direct Bellman recursion for this act-versus-wait margin and impose monotonicity only with respect to physical urgency, reflecting the irreversible loss of interception opportunity as time elapses. This structure enables early activation for dynamically demanding saves while preserving closed-loop adaptation when later observations change the predicted target. Under a single-crossing condition, MOS admits a threshold release boundary with a bounded approximation error. Extensive simulation studies show that MOS improves the mean save rate from 67.7% to 74.4% over a parameter-matched learned gate and increases reversal saves from 52.1% to 66.5%. Real-
robot experiments further demonstrate rapid interception and post-release direction correction under human shot-direction feints.
Read the full abstract ▾
All authors Hao E. Zhang, Ruize Geng, Yisen Li, Yaru Niu, Yikai Wang, Raihan Haque, Khalil Zbiss, Guanyang Luo, Hui-ping Wang, H. Eric Tseng, Ding Zhao
Artificial Intelligence
LEAP-NBV: Lightweight Edge Active-Perception for Foundation-Model Next-Best-View Planning
Boxun Hu, Jiawei Ge, Axel Krieger et al. (+2 more)
Foundation models are endowing autonomous systems with greater intelligence, enabling a more comprehensive understanding of the environment through visual perception. A representative example is Human Mesh Recovery (HMR), which provides useful estimates of a target's 3D pose and shape that can benefit tactical missions. However, the size and power demands of such models make them difficult to run on edge platforms and limit their real-time performance, undermining the requirements of tactical edge deployment - especially for active perception, where a mobile
robot must plan its next-best view on-board and cannot offload computation under contested communications. We present LEAP-NBV, a lightweight active-perception framework that runs foundation-model-driven Next-Best-View (NBV) planning on-board an edge device. To this end, we distill a family of large HMR teachers, each into a compact 32M student, with an offline mesh objective, then quantize the
vision encoder to FP16 and characterize its on-device accuracy and latency. Within an occlusion-aware active perception loop, we evaluate all configurations on the same held-out
benchmark and deploy the end-to-end pipeline on an NVIDIA Jetson Xavier NX, reporting measured on-device latency and energy.
Distillation recovers 6-7 mm of Procrustes-aligned mean per-vertex position error (PA-MPVPE) over the undistilled student on the test set. Selecting the edge-optimal compression model brings the HMR engine to ~12 ms at a small accuracy cost and runs the full closed loop at 3.6 FPS and 2.6 J per frame, achieving a 2.0x speedup and 3.0x lower energy than the uncompressed model while nearly matching downstream task quality.
Read the full abstract ▾
All authors Boxun Hu, Jiawei Ge, Axel Krieger, Peng Wang, Tinoosh Mohsenin
Systems and Control
Sufficient and Necessary Continuous Barrier-like Conditions for Discrete-Time Stochastic Reach-Avoid Verification
Bai Xue
This paper develops necessary and sufficient barrier-like characterizations using continuous barrier functions for infinite-horizon reach-avoid verification of discrete-time
stochastic systems. Existing results establish necessary and sufficient conditions in terms of functional inequalities involving measurable or lower semicontinuous barrier functions. However, the limited regularity of such functions may hinder their numerical approximation and computational synthesis. Building on our previous barrier-like condition for finite-horizon reach-avoid verification, we show that this condition can also be used for infinite-horizon reach-avoid verification and, under a uniform absolute continuity condition on the transition kernels, admits a continuous barrier function whenever the exact reach-avoid probability is strictly larger than the prescribed threshold for every state in the initial set. We further show that the resulting continuous barrier function can be uniformly approximated by a
polynomial one while preserving the required barrier-like conditions. For
polynomial systems, we formulate these conditions as
polynomial positivity constraints over compact basic semialgebraic sets. Putinar's Positivstellensatz then converts the positivity conditions into sum-of-squares (SOS) certificates, yielding semidefinite programming (SDP) formulations for synthesizing
polynomial barrier functions. We establish both soundness and completeness of the resulting SOS-based procedure. Finally, two numerical examples illustrate the theoretical results and demonstrate the resulting SDP approach.
Read the full abstract ▾
Artificial Intelligence
UniK: Universal Knowledge Perception for Digital and Physical AI
Nirmit Desai, Kunal Sawarkar, Aditya Mahakali et al. (+3 more)
Two transformative classes of AI systems are reshaping how organizations operate: digital AI, which reasons over enterprise knowledge to power chatbots and
agent workflows; and physical AI, which learns to control
robots and autonomous systems from
video, gameplay, and sensor telemetry. Both face the same foundational bottleneck: raw knowledge at scale, spanning heterogeneous modalities, locked in private corpora that existing AI infrastructure cannot access reliably or efficiently. We propose Universal Knowledge Perception (UniK) as a common platform for both classes, covering the full knowledge lifecycle (ingestion, enrichment, indexing,
retrieval, and continuous evaluation) across modalities from rich text and
video to molecular data and sensor telemetry. We present UniK, built on Polymath
Retrieval (multi-index fusion over automatically enriched indices) with no task-specific fine-tuning. Across five digital AI domains (medical literature, open-domain QA, chemistry, legal
video proceedings, and government open data) UniK combined with an open-source 70-billion-parameter model consistently matches or outperforms frontier proprietary LLMs that are orders of magnitude larger: 76% RAG accuracy on government data versus 47% for GPT-5; 77.9% on medical QA without fine-tuning; topping all open-source chemistry pipelines. We show that the same infrastructure directly addresses the data curation, indexing, and
retrieval challenges facing physical AI world model training, where the knowledge problem is harder but structurally identical.
Read the full abstract ▾
All authors Nirmit Desai, Kunal Sawarkar, Aditya Mahakali, Dongkon Lee, Kevin Park, Eric Song
Machine Learning · Cross-listed from stat.ML
Exponential Family Synthetic Controls
Hector Rodriguez-Deniz and David M. Blei
We develop exponential family synthetic controls (EFSC), a distributional version of synthetic controls for a panel of
datasets. Each
cell of the panel corresponds to a
dataset drawn from an exponential family whose natural parameters factorize probabilistically across units and times. We estimate the latent factors using black-box variational inference. This replaces the usual weighted-average view of synthetic controls with a flexible probabilistic model that operates on full distributions. We propose causal estimands based on divergences between pre- and post-intervention distributions induced by the posterior of the natural parameters, together with distributional placebo tests to support causal inference and assess the significance of the estimated effects. We validate the proposed framework on synthetic and real data. Across a variety of exponential-family distributions, EFSC accurately recovers causal effects induced by exponential tilts, together with the corresponding divergences between treated and counterfactual distributions. The framework also captures effects induced by structural perturbations of the latent factors and by heavy-tailed noise contamination. Finally, we apply EFSC to study the expansion of Medicaid under the Affordable Care Act (ACA) and its impact on the distribution of health insurance coverage across U.S. states. Code is available at https://github.com/blei-lab/efsc.
Read the full abstract ▾
Robotics
Opt2VLA: Force-Aware Vision-Language-Action for Contact-Rich Humanoid Whole-Body Manipulation
Fukang Liu, Yipu Chen, Jaehwi Jang et al. (+3 more)
Humanoid robots are expected to perform diverse human-level tasks in daily environments, many of which require precise regulation of interaction forces. While recent
vision-
language-action (VLA) models have shown promise for semantic planning and visuomotor control, existing
humanoid systems primarily represent actions through
geometric motion goals and rely on whole-body controllers focused on motion tracking, with limited explicit
reasoning or control of interaction forces. This limitation is particularly relevant in contact-rich tasks, where geometrically similar motions may require different force regimes depending on the task context and where visual observations may become unreliable after contact. In this work, we present Opt2VLA, a force-aware VLA framework that introduces explicit force commands at the VLA-to-control interface for
humanoid whole-body manipulation. A single multi-task VLA policy jointly predicts both
geometric motion goals and continuous contact-force references, which are tracked by task-specific reinforcement learning (RL)-based whole-body controllers. To provide scalable and physically grounded supervision, we generate dynamically feasible and contact-consistent training data via whole-body trajectory optimization (TO) with explicit force references. We evaluate Opt2VLA on three contact-rich
humanoid tasks and show that explicit force conditioning enables more accurate and consistent force regulation than motion-only control, while physically grounded torque supervision from TO further improves force tracking accuracy and stability. Closed-loop evaluations further demonstrate
language-conditioned force modulation with Opt2VLA in simulation and on
humanoid hardware.
Read the full abstract ▾
All authors Fukang Liu, Yipu Chen, Jaehwi Jang, Danfei Xu, Zsolt Kira, Ye Zhao
Computer Vision
Rethinking Diffusion Segmentation: When Does It Rely on Its Noisy State, and Does Diffusion Matter?
Hengzhuo Yang, Yuming Zeng and Yuling Yang
Diffusion models are increasingly adapted from generation to conditional prediction, where a conditioning signal is combined with an evolving noisy representation of the target. In fully supervised
segmentation, however, the conditioning image can already support direct target prediction, so endpoint performance alone establishes neither reliance on the added
diffusion state nor a deterministic advantage over image-only prediction. For state reliance, we disrupt target-derived state content or correct image-state pairing during retraining of twelve published methods across three
datasets, with ten matched seeds per setting. All 40 original-method comparisons whose evaluated-mask routes remained downstream of noised-quantity reconstruction exhibited state reliance, whereas all 30 comparisons with a
segmentation-supervised bypass preserved reference performance. Rerouting five originally bypass-capable methods by forcing
segmentation supervision through noise-to-mask reconstruction converted all 30 corresponding comparisons from preserved performance to state reliance. For deterministic utility, matched image-only counterparts achieved similar or better performance in 28 of 35 settings overall, including 16 of 20 whose native methods relied on both audited state properties. These results identify supervision path as a determinant of state reliance in the audited methods. Separately, matched image-only counterparts show that
diffusion-specific computation often provides no deterministic endpoint advantage, including in methods that rely on the audited state properties. More generally, when conditioning already supports strong target prediction,
diffusion-specific claims require additional evidence that the added state is used and that
diffusion-specific computation improves the claimed capability beyond a matched condition-only counterpart.
Read the full abstract ▾
Computation and Language
From Tables to Quantified Statements: Evaluating LLM Inference Generation through Executable Verification
Mai Mohamed Eida, Gunjan Anand, Ayush Singh et al. (+1 more)
LLMs can generate fluent descriptions from tables, but their outputs may remain logically unsupported by the structured data. We introduce STAT-TO-TEXT, a controlled task in which LLMs generate quantified natural
language inferences from statistical tables using quantified constructions such as all, some, no, and most. To evaluate these inferences, we use an LLM generated Python checker code which when executed verifies the corresponding truth conditions against the table. We compare four open-weight LLMs across model families and scales, evaluating faithfulness, logical accuracy, table coverage, and diversity. Our results show that model scale and family matter, with the largest model (GPT-OSS-120B) consistently producing the most faithful inferences without sacrificing greater table coverage and quantifier diversity, as opposed to smaller models. These findings are supported by human annotation, which shows that the automated checker closely aligns with human judgments.
Read the full abstract ▾
All authors Mai Mohamed Eida, Gunjan Anand, Ayush Singh, Aleksandre Maskharashvili
Computer Vision
Colon3R: Cross-Domain 3D Reconstruction from Monocular Colonoscopic Video
Zhihao Xing, Yingyu Wang, Liang Zhao et al. (+1 more)
Monocular colonoscopic 3D reconstruction is important for
surgical robotic colonoscopy, but remains challenging due to weak texture, specular reflections, limited view overlap, and non-rigid tissue motion. Conventional multi-view 3D reconstruction methods rely on stable correspondences and approximate
rigidity, which are often violated in colonoscopy. Existing endoscopic methods often rely on domain-specific supervision, whereas there are not enough in-vivo labeled data available to adapt geometry foundation models to clinical colonoscopy. We present Colon3R, a cross-domain semi-supervised framework built on pretrained VGGT that transfers coupled camera, depth, and pointmap geometry from labeled phantom and simulated data to unlabeled in-vivo colonoscopy without requiring target-domain
geometric annotations. Unlike source-only fine-tuning, which learns only from phantom and simulated data, Colon3R directly exploits unlabeled in-vivo
video through teacher-derived cross-view supervision. Our proposed hierarchical quasi-rigid reliability selects reliable supervision at the sequence, directed-pair, and pixel levels, while source-preserving adaptation retains the learned coupled geometry during target-domain adaptation. Extensive experiments demonstrate that our method achieves superior overall performance over state-of-the-art approaches in depth, pointmap, and camera pose estimation. Qualitative comparisons on real in-vivo colonoscopy further show substantially more complete and geometrically consistent reconstructions than competing methods under clinical domain shift. The code is publicly available at https://github.com/vglx/Colon3R.
Read the full abstract ▾
All authors Zhihao Xing, Yingyu Wang, Liang Zhao, Shoudong Huang
Computation and Language
Open-Jev Judgments on CallScreenBench: Calibrated One-Pass Scam Screening with a Small Language Model
Simiao Ren, Kidus Zewde, Xingyu Shen et al. (+6 more)
Screening a phone call for fraud needs a trustworthy probability after every caller turn, in milliseconds. Jev-style typed decisions promise exactly that: declared options go in, one calibrated probability per option comes out of a single forward pass, with no generated text. We test an open implementation of this readout, JevLite, on scam-call screening: Qwen3-4B is LoRA-tuned so that the temperature-scaled softmax over two answer-label logits is P(scam). On 41 held-out CallScreenBench scenarios (577 per-turn decisions) a three-seed ensemble reaches AUROC .974 with calibration error .052, non-inferior to an LLM judge (MiniMax-M3) at a pre-registered .02 margin, with no false alarms on legitimate calls, decisions 1.14 turns earlier under the same hang-up rule, and 64.5 ms per decision on one consumer GPU, 4.9x lower than the same backbone fine-tuned to generate its answer. The gain is in the readout and calibration, not accuracy: a fine-tuned ModernBERT encoder is not significantly worse, the recipe was selected with test-set exposure, and all callers are synthetic. We claim no architectural novelty; the contribution is the application and an evaluation reporting calibration, false alarms and decision timing alongside AUROC.
Read the full abstract ▾
All authors Simiao Ren, Kidus Zewde, Xingyu Shen, Yuchen Zhou, Dennis Ng, Ankit Raj, Tommy Duong, Yuxin Zhang, Neo Tiangratanakul
Human-Computer Interaction
When AI Tutors Speak: Evidence from a Randomized Field Experiment
Shihao Yang, Marshall Van Alstyne and Chrysanthos Dellarocas
Students increasingly study alongside generative artificial intelligence (AI), yet unguided access to fluent answers invites cognitive offloading, and there is little evidence on which configurations of AI tutoring produce learning. Two design margins are usually bundled together: pedagogical structure (how the tutor teaches) and interaction modality (how students talk to it). We separate them. In a preregistered randomized field experiment in a graduate corporate-finance course of an online MBA, we randomized 86 students between a structured tutor grounded in the course materials and a holdout in which consumer AI remained freely available. Within the tutored arm, each student's channel alternated weekly between voice and text, so the modality effect is identified within student. Structure mattered: tutored students gained 6.63 points more than ability-matched peers (p=.007), and the gain was concentrated in written
reasoning, where the share of answers reaching relational quality rose from 8% to 49% in the tutored arm against 8% to 27% in the holdout. Modality did not matter for learning. The instructor's own final, on file for all 86 randomized students, shows the same direction (2.6 points of 100, with no difference on a pre-treatment midterm). Voice nearly doubled conversational interaction and cost 2.8 times as much to deliver, yet it produced weekly mastery statistically equivalent to text, even as students came to prefer it. Pedagogical structure shapes what students practice, and modality shapes how they interact with the tutor. Making an AI more humanlike does not by itself make it more educational.
Read the full abstract ▾
Artificial Intelligence
Divergent strategies and convergent outcomes in autonomous materials discovery
Jihan Kim
Scientific
agents are mostly evaluated on whether they complete tasks or recover known results; we instead study variation across repeated open-ended campaigns. Sixteen separately initialized sessions of one model-harness configuration received a frozen database of 12,499 metal-organic frameworks, a methane-storage objective, a pinned protocol and a one-week budget. Strategies diverged into four approaches spanning 100–5,000 screened structures, and eight built 2,253 hypothetical structures. Yet the
agents recovered the same materials frontier near 200 cm^3/cm^3, and an independent calculation of the database's porous region found its nine best structures all among their reports. Enforced checks on half the
agents raised fresh-run reproduction from one of eight to eight of eight but could not detectably improve conclusion validity, because fifteen of sixteen
agents selected the same audit-excluded entry, an incomplete structure whose missing anions created artificial pore volume. Replicated
agents thus reveal both robust conclusions and common-mode errors from shared inputs.
Read the full abstract ▾
Computation and Language
Some Dialects Are More Equal Than Others: Non-Prestigious Arabic Dialectal Bias in LLMs
Mai Mohamed Eida, Ryan Dolan, Paul de Nijs et al. (+1 more)
Previous work on Egyptian Arabic in NLP has focused largely on the prestigious Cairene Egyptian Arabic (CEA) dialect, resulting in a lack of representation for the less prestigious Sa'idi Egyptian Arabic (SEA) dialect both in LLM and resource development. Does this lack of representation influence an LLM's view of the acceptability of SEA (upstream), and does an upstream bias against SEA lead to worse performance (downstream)? We investigate the upstream effect of SEA dialectal features on LLM preferences in a Targeted Syntactic Evaluation (TSE) task which reveals a significant bias against SEA across multiple LLMs. We then analyze the effect of these same features on downstream model performance on MMLU
benchmarks and show that models experience a degradation in performance when presented with SEA. This work highlights the need for further exploration on how sub-dialectal variation impacts
language technologies.
Read the full abstract ▾
All authors Mai Mohamed Eida, Ryan Dolan, Paul de Nijs, Jonathan Dunn
Programming Languages
Djinnlang: Higher-Level Programming by Unambiguous Specification with an LLM in the Compiler
Simon Henniger, Stephen Chong and Nada Amin
Programmers write formal specifications, and LLMs implement them, proving that each implementation matches its spec. Taken to its extreme, this makes specification
languages the new programming
languages. We argue that an unambiguity constraint is key: in addition to proving that its implementation satisfies the specification, the LLM must also prove that any other implementation satisfying it must produce the same outputs on the same inputs, i.e. that the relation formed by the constraints is deterministic. This leaves the LLM no leeway on program semantics: as with a conventional compiler, the generated code never needs to be read and can be regenerated from the spec at any time. Under this constraint and with a powerful LLM, the difference between a specification
language and a programming
language becomes essentially meaningless, and the LLM essentially becomes a part of the compiler toolchain. The arrangement doubles as a strong form of AI control: an untrusted model writes the code, yet its work is tightly checked by a verifier. To demonstrate that our LLM-in-the-compiler paradigm is feasible when supported by our unambiguity constraint, we present Djinnlang, a high-level specification
language built for this future. A Djinnlang program consists only of specifications — the programmer never writes executable code. In place of a traditional compiler, a symbolic translator lowers each spec to Dafny stubs and proof obligations, and a driver harness orchestrates an LLM that fills in implementations and proofs, all checked by the Dafny verifier. We evaluate our
language and implementation on multiple examples and we show that it is self-hosting: an LLM can implement the Djinnlang translator from its specification and the reimplementation can verify itself.
Read the full abstract ▾