Machine Learning
Cross-Benchmark Transfer from RL on Agentic Coding Tasks
Sushant Mehta, Logan Ritchie and Edwin Chen
Coding
agents often fail in the last mile: they build most of a feature but drop a requirement, test only the cases their implementation already handles, break behavior that was supposed to stay intact, or validate against an unchecked assumption. We ask whether reinforcement learning (RL) on expert-built agentic coding tasks closes this gap, and whether what the
agent learns transfers beyond the training distribution. We post-train Kimi K2.7 Code, a 1T-parameter (32B active) open-weight mixture-of-experts model, with RL alone on 1,700 tasks: 1,000 repository tasks graded by hidden fail-to-pass tests and by pass-to-pass tests of existing behavior, and 700 terminal tasks graded by expert-written hidden verifiers. The reward is the fraction of target checks passed and drops to zero if any pass-to-pass test fails. One epoch of GSPO on a rank-32 LoRA adapter improves pass@1 on each of the six external
benchmarks we evaluated, across three
agent harnesses: SWE-Bench Pro (60.1 to 64.8), DeepSWE (31.0 to 43.4), Terminal-Bench 2.1 (67.4 to 82.0), Terminal-Bench 3 (1.4 to 12.1), Terminal-Bench 4 (0.0 to 7.6), and SWE-Marathon (5.0 to 25.0). Pooled over the five independent task sets (Terminal-Bench 4 revises Terminal-Bench 3), the improvement is significant (p < 0.001), and it remains significant on the three sets released after the training data was collected (p = 0.004); the model also improves under both harnesses never used in training. Median trajectories on DeepSWE and Terminal-Bench 3 are 24-35% shorter in
agent steps. The base model's failed DeepSWE runs are mostly near-misses, and on the tasks the trained model newly solves, paired trajectories show it avoiding each of the four failure modes above.
Read the full abstract ▾
Databases
HakiCC: LLM-Driven Multi-Agent Design and Optimization of Concurrency Control Protocols
Farzad Habibi, Juncheng Fang and Faisal Nawab
Large
language models (LLMs) have recently been applied in systems research as a tool to reduce human-intensive engineering effort through cost-efficient automation. Decades of research have produced a rich landscape of concurrency control (CC) protocols, each encoding distinct trade-offs in correctness, throughput, and abort behavior. However, most applications in practice default to 2PL or OCC, because selecting and adapting a protocol to a specific application requires expert knowledge that is rarely available to application designers. This is a wasted opportunity, as an application-specific CC protocol can yield significant performance advantages over a generic baseline, but designing one requires deep expertise in CC protocol design. In this paper, we propose HakiCC, an LLM-driven multi-
agent pipeline that automatically designs, verifies, and optimizes concurrency control protocols tailored to a given target application. HakiCC provides a two-stage pipeline. In Stage 1, a multi-
agent system takes a workload description as input and generates an application-specific CC protocol implementation, which is iteratively repaired and verified for conflict-serializability. In Stage 2, the verified protocol is further optimized for that application through an LLM-driven evolutionary loop targeting correctness and throughput. We evaluate HakiCC on TPC-C and AuctionMark as target workloads, producing and reporting ten application-specific CC protocols. All ten are conflict-serializable after Stage 1; Stage 2 improves throughput for every protocol, with average gains of +50.6% for TPC-C protocols and +92.2% for AuctionMark protocols.
Read the full abstract ▾
Machine Learning
Match the Distribution, Not the Compute: Post-Training Multi-Token Prediction Heads
Prachi Badarayani, Aidan Jay, Chenghui Zhou et al. (+8 more)
Multi-
token prediction (MTP) improves the throughput of autoregressive generation by enabling the
language model to draft multiple next
tokens per forward pass, while a verification step over draft
tokens ensures that
token distribution of the backbone is preserved. Every open MTP-family release (
MiMo-7B, DeepSeek-V3, Qwen3) trains its heads jointly with the backbone over the full pretraining run of tens of trillions of
tokens, thus setting the drafter quality at pretraining time. We ask whether a lightweight post-training pass on target-generated chain-of-thought is enough to reach the same expected throughput speedup on a frozen
reasoning model, and study how a serving-time system built on such a checkpoint can be optimized. We present three findings. 1) On a frozen Qwen3-8B with $K{=}3$ chained MTP heads, we show that a post-training recipe with plain cross-
entropy on $\approx\!2.5$B
tokens reaches or exceeds the expected speedup of jointly trained
MiMo-7B on math, coding and knowledge
benchmarks. Our post-training recipe utilizes $10^3$-$10^4\times$ less MTP-training
tokens as compared with joint pre-training of
MiMO-7B MTP baseline. 2) We propose a chain-aware relaxation of draft
token verification rule that allows a bounded drift from backbone
language model
token distribution. We show that this relaxation lifts expected speedups by $+12$ to $+16\%$ per
benchmark while preserving task accuracy. 3) We propose an adaptive controller that dynamically chooses the number of MTP heads to be engaged at inference time and demonstrate recovery of upto $11$–$14\%$ loss in speedup using fixed maximum MTP draft length.
Read the full abstract ▾
All authors Prachi Badarayani, Aidan Jay, Chenghui Zhou, Dayquan Julienne, Yuan Gao, Tianwei Chen, George Zerveas, Ishmam Zabir, Xiren Zhou, Chris Quirk, Xia Song
Software Engineering
FORALL-LEAN-AGENT for Auditable Reasoning in Formal Mathematics and Software Verification
Naing Oo Lwin
Coding
agents increasingly automate Lean proof development, but successful compilation alone does not establish that a candidate proves the intended statement under acceptable assumptions. We present FORALL-LEAN-
AGENT, a frontend-agnostic framework for auditable
reasoning in formal mathematics and software verification. The framework combines isolated workspaces, Lean tools, and fresh review with statement comparison, axiom audits, and independent proof checking where supported. Verification evidence and reviewer decisions are bound to the same candidate artifact, making acceptance traceable. We evaluate the framework on VeriSoftBench, PutnamBench, and both problems in the Lean Eval softwareverification track. On the 100-task VeriSoftBench subset, integration with FORALLLEAN-
AGENT raises
benchmark-rule success from 93 to 100 for GPT-5.6 Sol at low effort while reducing cost from $69 to $62. The PutnamBench evaluation accepts all 672 problems at an average of $4.72 each. These results show that
agent harness design can improve correctness and efficiency while providing evidence beyond aggregate solve counts.
Read the full abstract ▾
Computation and Language
DeBERTa-ConPara: Attack-Aware and Deployment-Realistic Detection of AI-Generated Text
Mohamed Mady, Yupei Li, Johannes Reschke et al. (+1 more)
Robust detection of AI-generated text under deployment conditions is challenging: distribution shifts across domains and generators,
adversarial perturbations of the input surface, and the absence of target-domain labels for threshold calibration all degrade detectors that perform well in-domain. We present DeBERTa-ConPara, a deployment-oriented detector combining attack-aware Unicode preprocessing with a contextual
transformer encoder trained over HC3 Plus, M4, MAGE and RAID. Our central finding is that preprocessing acts in opposite directions depending on where it is applied: normalising the training corpus deduplicates it, collapsing 35.4% of RAID rows into copies of their clean siblings and deleting the
adversarial supervision, whereas normalising at inference is an effective defence. A factorial varying the two placements independently identifies raw training with normalised inference as the best configuration, reaching 99.61% AUROC, 99.01% TPR@5% FPR and 96.57% TPR@1% FPR on the official RAID hidden test, alongside 93.14% average balanced accuracy across HC3 Plus and MAGE under a fixed threshold. The gain is confined to two of twelve attack classes: homoglyph and zero-width-space insertion rise from 11.05% and 1.12% to 96.98%. The same signature reproduces in a zero-shot detector of different architecture, showing the effect belongs to the attacks rather than to our model. We additionally report two negative results: semantic-invariance augmentation through paraphrasing and supervised contrastive learning (ConPara) does not improve the best configuration, and the handcrafted feature-fusion branch is inert in distribution and harmful outside it.
Read the full abstract ▾
All authors Mohamed Mady, Yupei Li, Johannes Reschke, Björn W. Schuller
Numerical Analysis
Uncertainty Quantification for Landweber Iteration with Randomized Normal-Operator Approximation
Anuj Abhishek and Sean Holman
Iterative methods are extremely popular for solving linear ill-posed inverse problems. While deterministic convergence, or more precisely, semi-convergence of such methods is widely studied, the problem of quantifying uncertainty in the resulting reconstructions due to random noise in the observed data has received far less
attention. In this work, we study uncertainty quantification for the Landweber iteration in linear inverse problems using the Radon transform as a motivating example. We interpret the Landweber reconstruction statistically by analyzing how uncertainty in the measured data is propagated through the reconstruction map. This perspective is closely related to generalized fiducial inference, where uncertainty about the parameter is induced by inverting the relation between the observed data and the unknown (fixed) quantity. We combine the resulting
stochastic uncertainty with a theoretical bound on the regularization bias to construct confidence intervals for the reconstructed solution. The evaluation of the resulting uncertainty estimates require repeated operations with the forward operator that can become expensive in large-scale problems. To reduce this cost, we use randomized singular value decomposition to obtain a low-rank approximation of the normal operator associated with the Radon transform and incorporate this approximation in quantifying uncertainty. {{In particular, this requires that the additional randomization error introduced due to the use of such randomized techniques be included in our approach for uncertainty quantification.}} Numerical results show that the proposed method provides accurate reconstructions and reliable uncertainty estimates at substantially reduced computational cost.
Read the full abstract ▾
Computer Vision
Machine Translation for Sign Languages
Ozge Mercanoglu Sincan, Anton Pelykh, Edward Fish et al. (+7 more)
Sign
language machine translation has progressed substantially over the past decade, evolving from isolated sign recognition to end-to-end translation systems. Advances in pose estimation,
transformer architectures, and large-scale
dataset collection have driven progress, yet challenges remain.
Datasets are limited compared to spoken-
language resources; evaluation metrics inadequately capture the linguistic quality of output; and models must capture the simultaneous, multi-layered, and three-dimensional structure of sign
languages. This manuscript provides a comprehensive review that seeks to balance technical challenges with stakeholder considerations. We examine the linguistic properties that make sign
languages computationally unique, trace the
evolution of recognition, translation, and production systems, and analyze ongoing technical challenges. Crucially, we address ethical considerations around data governance, community involvement, and appropriate use. Drawing on interdisciplinary perspectives spanning computer
vision, sign
language linguistics, and deaf studies, our analysis emphasizes that continued progress requires sustained collaboration across these fields and with deaf communities.
Read the full abstract ▾
All authors Ozge Mercanoglu Sincan, Anton Pelykh, Edward Fish, Harry Walsh, JianHe Low, Karahan Sahin, Oline Ranum, Sobhan Asasi, Steven Emery, Richard Bowden
Robotics
UniTrackPLA: Unified Panorama-Language-Action Model for Instruction-Guided Navigation and Dynamic Person Tracking
Pengfei Qi, Haoran Lin, Sizhuang Chen et al. (+7 more)
General-purpose
embodied robots should support both
navigation toward
language-specified destinations and dynamic person tracking under arbitrary initial target azimuths. However, existing methods typically rely on forward-facing observations and address these tasks with separate policies, limiting omnidirectional perception and unified closed-loop control. We present UniTrackPLA, a unified panorama-
language-action model for instruction-guided
navigation and dynamic person tracking. Its Panoramic-Aware Encoding (PAE) preserves the temporal and azimuthal structure of perspective views projected from each panorama, enabling perspective-pretrained visual encoders to process omnidirectional observations. A shared
vision-
language backbone grounds instructions in the panoramic context and predicts continuous
robot-centric waypoint chunks for both tasks. World-Action Consistency (WAC) further predicts action-conditioned future visual states and verifies waypoint prefixes online, allowing reliable actions to be reused while triggering replanning upon inconsistency. We also introduce OmniTrackNav-Bench, comprising 5,000 simulated tracking trajectories, 10,000 simulated VLN routes, and 96 verified real-world routes, providing 919,978 waypoint-supervision instances. UniTrackPLA improves overall tracking SR from 23.50% to 35.00% and Omni-VLN SR/SPL from 13.00%/12.77% to 19.75%/19.29%. Incorporating 76 real-world routes further improves held-out EP@0.2m from 42.92% to 92.08%. Closed-loop experiments on a Go2-W
robot demonstrate unified panoramic tracking and
navigation across indoor and outdoor environments. The project page is at https://tw5775.github.io/UniTrackPLA.
Read the full abstract ▾
All authors Pengfei Qi, Haoran Lin, Sizhuang Chen, Kai Luo, Sirui Zhang, Xinqi Liu, Fei Cheng, Wenrui Chen, Liming Yin, Kailun Yang
Data Structures and Algorithms
Beyond odd characteristic: Faster isomorphism testing of 2-groups of Frattini class 2
Joshua A. Grochow, Gábor Ivanyos, Youming Qiao et al. (+1 more)
The finite group isomorphism problem asks whether two finite groups of order $N$ are isomorphic. The first
algorithm, attributed to Tarjan (see Miller, STOC '78), runs in time $N^{\log N + O(1)}$. Despite intensive study, the current best known
algorithm has a running time of $N^{(1 / 4 + o(1))\log N}$ (Rosenbaum, '13). $p$-groups of class $2$ have been recognized as the major bottleneck for faster group isomorphism. Recent progress has led to $N^{o(\log N)}$-time
algorithms for $p$-groups of class $2$ where $p$ is odd (Sun, STOC '23; Ivanyos–Mendoza–Qiao–Sun–Zhang, FOCS '24; Grochow–Qiao–Stange–Sun, STOC '25). However, the case of $p=2$, which represents the majority of $p$-groups of class 2 assuming a well-known
conjecture in group enumeration, remained elusive, with essentially no progress until now. In this paper, we present an
algorithm for testing the isomorphism of two 2-groups of Frattini class 2 of order $N$ in time $N^{O((\log N)^{1/2})}$. To our knowledge, this is the first $N^{o(\log N)}$-time isomorphism
algorithm for a class of $2$-groups that constitutes logarithmically almost all $2$-groups, in the sense that $\lim_{N \to \infty} \frac{\log(\text{\# 2-groups of Frattini class 2 and order } \leq N)}{\log(\text{\# 2-groups of order} \leq N)} = 1$. As our main tool, we present the first non-trivial
algorithms for the quadratic form space/tuple isometry problems over $\mathbb{F}_2$. These
algorithms rely on combinations of combinatorial and algebraic ideas, including finite
matrix group
algorithms developed by Luks (FOCS '92). As far as we know, this is the first time that
matrix group
algorithms are used to make progress on the worst-case complexity of $p$-group isomorphism.
Read the full abstract ▾
All authors Joshua A. Grochow, Gábor Ivanyos, Youming Qiao, Xiaorui Sun
Machine Learning
Rethinking Data Augmentation under Covariate Shift: Invariant-Guided Diffusion and Prototype Reweighting
Hongyu Cao, Xinyuan Wang, Arun Vignesh Malarkkan et al. (+3 more)
In many industrial applications, 1) tabular data is scarce and imbalanced and thus requires synthetic expansion; 2) input distributions drift between training and deployment (
covariate shift); 3) validation sets often diverge from unseen test environments; or 4) standard generative models simply mimic outdated source distributions. This learning setting limits the stability of standard augmentation and adaptation pipelines. We generalize the task under such setting as the Augmented and Weighted Learning under
Covariate Shift problem (AWL-CS). AWL-CS imposes two critical challenges on existing methods: 1) misleading generative guidance where models optimize for source similarity rather than downstream task relevance, and 2) structural instability of distributional density where reweighting mechanisms overfit to noisy validation signals. To tackle these challenges, we propose IGDPR (Invariant-Guided
Diffusion with Prototype Reweighting), a unified framework that synergizes stable synthesis and structural adaptation: i) To achieve task-relevant generation, we steer the
diffusion sampling process using invariant potentials to ensure synthetic samples align with stable decision boundaries rather than outdated correlations. ii) To ensure stable adaptation, we develop a prototype-based reweighting strategy that assesses sample reliability through structural clusters instead of isolated points, effectively filtering validation noise. Extensive experiments on real data demonstrate our method improves data quality by augmenting the most beneficial data for robust learning.
Read the full abstract ▾
All authors Hongyu Cao, Xinyuan Wang, Arun Vignesh Malarkkan, Kunpeng Liu, Haifeng Chen, Yanjie Fu
Artificial Intelligence
MemFit: Efficient Long-Term Agentic Memory
Mitchell Piehl and Muchao Ye
Long-term
memory systems for large
language models (LLMs) have gained popularity for extending
reasoning capabilities across applications. Current
memory systems rely on LLM
agents to organize and consolidate
memory, resulting in costly, inefficient write operations. To address this limitation, we propose MemFit, a long-term
memory system for conversational
agents that reduces the cost and latency of
memory operations. Unlike existing systems that rely on expensive LLM calls for
memory construction or discard surface-level details through compression, MemFit stores each turn verbatim in an append-only store with near-instantaneous, LLM-free insertion, indexing turns with segment summaries rather than replacing them. Additionally, MemFit uses an LLM-free, multi-path
retrieval strategy that combines lexical and semantic signals with cross-encoder reranking over caption- augmented episodes in both textual and
multimodal settings. Empirical results on three widely used
benchmarks, LoCoMo, MemGallery, and LongMemEval-S, show that MemFit achieves state-of-the-art performance while reducing
memory construction time and cost several-fold, providing a scalable and efficient solution for persistent agentic
memory.
Read the full abstract ▾
Artificial Intelligence
An Educator-Guided LLM Pedagogical Agent for Scaffolded Feedback in Conceptual Database Design
Sara Riazi and Pedram Rooshenas
We present an educator-guided LLM pedagogical
agent for scaffolded feedback in conceptual database design. Integrated into an entity–relationship diagram (ERD) editor, the system grounds feedback in the student artifact, assignment requirements, educator-authored rubrics, and instructional resources. Its architecture separates hidden, artifact-grounded diagnosis from the workflow that controls the form and disclosure level of student-facing support. We instantiate the architecture as a four-stage workflow progressing from concept checks and guided application to low-detail feedback and localized clarification. Each feedback request creates a stateful episode linked to versioned ERD states. In a deployment spanning three ERD environments and 383 feedback episodes, 71.1% of observed target-level changes fully or partially incorporated the hidden diagnostic target, including many after Stages~1–2. Qualitative analysis showed that staged disclosure sometimes withheld inaccurate details, supported selective uptake, or allowed later recovery, though some errors still shaped revisions. Survey responses from a self-selected sample favored delayed disclosure and student agency but noted indirectness and repetition.
Read the full abstract ▾
Robotics
Kinematic MeanFlow: One-Step Action Generation Policy for Robotic Foundation Models
Jiawei Fan, Sifeng Wang, Yuqing Hou et al. (+1 more)
In this paper, we study how to achieve one-step action generation in Robotic Foundation Models (RFMs), aiming to overcome the high inference latency of multi-step flow matching. MeanFlow provides a promising framework for this goal, yet its direct application leads to performance collapse. We discover that this stems from two distinctive dynamics exhibited in the RFM velocity field: (1) the “local acceleration" exhibits stability early on, but surges sharply towards the end of the denoising process, and (2) the spread of its magnitudes across samples widens as denoising progresses. To address these issues, we introduce Kinematic MeanFlow (K-MF), a novel one-step action policy tailored for RFMs. Specifically, grounded in a kinematic identity, K-MF decouples the time derivative term in the MeanFlow formulation into two sub-interval terms separated by an intermediate point. This decoupled formulation enables the two terms to capture early-stage and late-stage denoising dynamics, respectively, while mitigating the error amplification across the process. As a result, our K-MF empowers RFMs to achieve one-step action generation in both training from scratch and fine-tuning paradigms across diverse tasks, while outperforming multi-step flow matching in most settings. In terms of inference efficiency, K-MF reduces action-head latency of GR00T-N1.6 by 67.5%~74.4% across L40 and Jetson Orin in eager and compiled modes, yielding end-to-end latency reductions of 30.3%~54.9%. Code will be available at https://github.com/IntelChina-AI/K-MF.
Read the full abstract ▾
All authors Jiawei Fan, Sifeng Wang, Yuqing Hou, Anbang Yao
Computers and Society
Correctness, Convergence, and AI-Generated Code Detection: A Longitudinal Study of Student and Large Language Model Code in Introductory Programming
Runlong Ye, Jing Fan, Angela Zavaleta Bernuy et al. (+4 more)
Large
language models can generate plausible solutions to programming assignments, making it tempting to detect their use by matching student code against a reference bank of generated solutions. Yet similar code can also arise when an assignment admits only a few natural implementations, which leaves open what a match actually shows. We investigate generated-reference matching using 29,970 student submissions from ten Python labs offered in 2021, 2023, and 2025, together with 90,000 solution attempts generated retrospectively by three frontier LLMs. We validate the generated solutions using hidden instructor tests, compare code with MOSS after excluding the starter code, and examine the exact abstract syntax tree (AST) forms of selected functions. The models usually produced correct solutions and, across most assignments, converged on similar implementations. Student submissions matched the generated references more often in later cohorts, including among submissions that passed every hidden test. On tightly specified functions, the models converged on a few exact abstract-syntax-tree forms, and the number of distinct student forms also declined across cohorts, whereas open-ended functions remained diverse in both sources. Most reported overlaps were short, making the minimum match length an important choice when reviewing students' code. Finally, we discuss how instructors can build a reference bank of generated solutions before releasing an assignment to identify tasks on which generated solutions converge, decide how much review a match warrants, and redesign tasks to elicit tests,
reasoning, and intermediate work. These findings support tracking population-level changes in submitted code, while attributing AI use to an individual submission would require additional evidence about how it was produced, such as
prompts, revisions, intermediate code, and student disclosures.
Read the full abstract ▾
All authors Runlong Ye, Jing Fan, Angela Zavaleta Bernuy, Oscar Karnalim, Paul Denny, Juho Leinonen, Michael Liut
Machine Learning
Don't Waste the Noise: Importance-Guided Perturbation Allocation under Joint Global and Local Constraints
Melika Shirian and Kianoosh Vadaei
Adversarial optimization under a shared $\ell_1$ budget requires deciding not only how much perturbation to use, but also where that limited budget should be spent. This allocation problem becomes particularly important when individual input coordinates are subject to local magnitude constraints, which restrict the extent to which perturbation can be concentrated on a small number of locations. We introduce an importance-guided allocation mechanism that uses a fixed clean-gradient prior to steer perturbation toward model-sensitive regions while leaving the feasible perturbation set unchanged. A centered allocation objective encourages perturbation at above-average importance locations and discourages unnecessary expenditure elsewhere, thereby redistributing rather than enlarging the available budget. Across ten robust model–
dataset configurations under a common capacity-limited threat setting, the proposed method improves attack success over matched APGD- and PMA-based baselines by $2.52$ to $17.70$ percentage points. Allocation analysis shows that these gains are accompanied by substantially greater perturbation mass in high-importance regions without increased global $\ell_1$ consumption. Mechanism ablations further show that centered non-uniform redistribution provides part of the benefit, while model-derived importance yields an additional improvement. These results identify perturbation allocation as a distinct and practically relevant dimension of
adversarial optimization under shared-budget, locally constrained threat models.
Read the full abstract ▾
Computer Vision · Cross-listed from eess.IV
MorphoBranch: A Fine-Structure-Preserving Workbench for Morphometric Analysis of Branched Cellular Structures
Song Zhiying, Ling Hanyi, Wu Junyi et al. (+1 more)
Background and Objectives: Fluorescence-labeled cellular arbors provide readouts of neuronal and microglial morphology, but fine and weakly labeled processes are prone to fragmentation and false connections that bias skeleton-based measurements. We present MorphoBranch, a fine-structure-preserving, human-reviewable workbench for morphometry of branched cellular structures. Methods: MorphoBranch combines a deterministic Morphometry Engine with an LLM-assisted Refinement Engine. The Mor- phometry Engine implements an image-to-graph workflow integrating multiscale structural evidence extraction, hysteresis segmen- tation, evidence-constrained skeleton refinement, and graph-based morphometry. The Refinement Engine maps natural-
language requests to registered actions for parameter adjustment, preview execution, metric reporting, and unsupported-request handling, while image processing and quantitative computation remain deterministic and reviewable. Results: MorphoBranch was evaluated on two public neuronal axon
datasets, AxonMIP and AxonStack, and the in-house
Cell- Morph
dataset of microglial fluorescence images. It achieved the highest Skeleton F1 and clDice and the lowest length-estimation error among the evaluated methods on all three
datasets, while also achieving the highest Dice and IoU on AxonMIP and Axon- Stack. Across 150 natural-
language tasks, the Refinement Engine achieved a 94.0% end-to-end success rate. Conclusions: These results demonstrate that MorphoBranch provides a reproducible, human-reviewable workflow for mor- phometric analysis of branched cellular structures. It supports fine-structure-preserving quantification across neuronal axon and microglial fluorescence images while maintaining inspectable and reproducible analysis workflows.
Read the full abstract ▾
All authors Song Zhiying, Ling Hanyi, Wu Junyi, Jiang Yangbo
Computer Vision
CtrlWAM: Controllable World Action Models with Aligned Intent and Foresight
Chensheng Peng, Wenhao Ding, Ran Tian et al. (+9 more)
World action models (WAMs) jointly predict actions (intent) and visual future (foresight). Standard training adds noise to recorded actions and
video simultaneously, but such training paradigms introduce a mismatch: perturbed actions imply counterfactual future visual, while the noised
video remains tied to the GT recording. In low-noise regime, the scene geometry and even the dynamic behavior remain clearly visible from the noisy future frames despite the added noise. We present CtrlWAM, which executes perturbed actions in a simulator and pairs them with their noised visual consequences for joint WAM learning. To accommodate the different denoising requirements of
video and actions, we introduce warped
video–action noise schedules that aim to keep visual layout responsive as action predictions evolve. We further extend the action interface from ego-only control to a variable number of
agent streams, allowing a unified model to represent predicted or commanded futures for multiple
agents. Driving experiments show more accurate action forecasts, closer agreement between generated
video and actions, and better following of supplied commands; robotics experiments show stronger motion fidelity and controllability. Matched controls support the benefit of off-path renders for command following and manipulation fidelity. Together, these findings contribute to a more controllable world action model. Project page: https://ctrl-wam.github.io/
Read the full abstract ▾
All authors Chensheng Peng, Wenhao Ding, Ran Tian, Zewei Zhou, Jef Packer, Maximilian Igl, Peter Karkus, Yan Wang, Masayoshi Tomizuka, Boris Ivanovic, Marco Pavone, Yuxiao Chen
Machine Learning
Fixing a Model That Learned Worse Cancer Means Lower Risk: Monotonic Constraints in Bladder Cancer Recurrence Prediction
Saram Abbas, David Thomas, Naeem Soomro et al. (+3 more)
Background and Objective: Clinicians expect recurrence risk to climb with cancer severity. In a UK multicentre trial, an unconstrained XGBoost model learnt that higher tumour stage and carcinoma in situ predicted lower recurrence risk, and discrimination, calibration, and SHAP were all blind to it. We developed a counterfactual testing framework to detect this inversion and a monotonic-constraint framework to remove it without hurting performance. Methods: BOXIT enrolled 472 patients with protocol-mandated cystoscopy across 51 UK sites (2007-2012); 435 had at least two years' follow-up (153 recurrences, 35.2%). We developed a counterfactual direction test and a monotonic-constraint correction, with constraint directions drawn from the EORTC and EAU risk systems, and evaluated both against unconstrained XGBoost and logistic regression on 18 predictors (seven directed) over 50 cross-validation folds. The test worsened each patient on one directed feature at a time to check whether risk fell; SHAP direction and calibration were also assessed. Key Findings and Limitations: Tumour stage and carcinoma in situ were associated with lower recurrence, opposite to medical intuition; the unconstrained model reversed carcinoma in situ counterfactuals in 90.2% of cases and stage in 74.3%. Discrimination ($\Delta$AUC 0.005, p=0.47), calibration, and SHAP magnitude were all blind to the inversion. Monotonic constraints eliminated every violation at no cost to discrimination (0.723 vs 0.718) and outperformed EORTC (p=8.9e-16). Limitations: single trial, internal-external validation only. Conclusions and Clinical Implications: A model that had learned this inversion passed every conventional check. A counterfactual direction test, run as a single refit with pre-specified monotonic constraints, catches this failure at no cost to performance and should be routine before clinical deployment.
Read the full abstract ▾
All authors Saram Abbas, David Thomas, Naeem Soomro, Rishad Shafik, Rakesh Heer, Kabita Adhikari
Computer Vision
Lang3DSeg: Annotation-Free Open-Vocabulary 3D Segmentation with Point Transformers
Cigdem Kokenoz, Amir Salarpour, Alkim Domeke et al. (+5 more)
Accurate 3D semantic perception is critical for safe autonomous
navigation. However, supervised
LiDAR segmentation remains tied to closed taxonomies and to the cost of point-wise manual annotation. Open-vocabulary methods avoid that cost by projecting the output of 2D
vision-
language models onto
LiDAR and distilling it into a 3D network. These methods rely almost exclusively on voxel-based sparse convolutions, and point
transformers have so far been limited to indoor environments, where 3D data is dense and bounded. We present Lang3DSeg, which establishes a point
transformer as the backbone for annotation-free open-vocabulary
segmentation of outdoor 3D
LiDAR, and is trained from scratch without
geometric pre-training. This training paradigm necessitates addressing the inherent noise in 2D-to-3D label projections; specifically, naive projection often suffers from depth ambiguity, where points behind an object are erroneously assigned its semantic label. We therefore composite masks using an explicit class-priority rule and truncate each projected instance at the first gap in its depth distribution, correcting the projection error directly rather than averaging it over registered sequences. Lang3DSeg achieves 52.8% mIoU on nuScenes validation and 41.4% on SemanticKITTI, the highest among published annotation-free methods on both
benchmarks. Every 3D semantic
segmentation is on a single
LiDAR sweep, and inference operates in real-time without running
vision-
language models.
Read the full abstract ▾
All authors Cigdem Kokenoz, Amir Salarpour, Alkim Domeke, Christopher Salas, Pedram MohajerAnsari, Long Cheng, Mert D. Pesé, Bing Li
Robotics
Are Frontier VLM Agents Ready to Be Robot Generalists? An Empirical Study with the Embodied Agent Arena
Haojian Huang, Pukun Zhao, Zexi Li et al. (+7 more)
A
robot generalist must interpret a scene, act on it, and satisfy every condition of a task. How far do frontier
vision-
language models (VLMs) meet these demands? We introduce
Embodied Agent Arena, evaluating ten VLM
agents across Geometry, Spatial
Reasoning, Affordance, Task Planning, and Manipulation. Its 1,000 cases combine 32 established sources with GeoProbe, our 168-case
benchmark for camera, motion, depth, and scale estimation. A minimal harness leaves task decisions to the model through source-native interfaces. Closed-weight models lead all five domains, whereas open-weight VLMs show useful spatial
reasoning but substantially lower planning and manipulation success. Beneath these averages, similar
geometric errors conceal different scale and shape distortions, and valid region estimates need not identify usable action points. During execution,
agents can achieve early subgoals yet lose object control, bind an action to the wrong receiving region, or stop before the required physical state holds. Additional computation does not uniformly resolve these failures: code repair can restore execution without completing the task, while self-review can reverse correct judgments. Failure costs also differ: the strongest planners concentrate failed-
token expenditure in prolonged body-control tasks, whereas other models fail broadly across procedures. Matched traces show how successful recovery uses feedback to restore a missing prerequisite and advance the remaining task.
Read the full abstract ▾
All authors Haojian Huang, Pukun Zhao, Zexi Li, Yehang Zhang, Yangkai Wei, Wenqian Li, Han Yang, Kaiwen Zhou, Ying-Cong Chen, Yinchuan Li
Computation and Language
Child-Adapted Structured Phonological Representations for Interpretable Speech Sound Analysis
Abner Hernandez, Tomás Arias Vergara, Andreas Maier et al. (+1 more)
Structured phonological representations provide an interpretable alternative to generic
speech embeddings, but existing models are largely trained on adult
speech. We adapt PhonoQ-2.0 to child
speech using CHILDES-Aligned data and compare three
alignment-supervision conditions (Adult, Adult+Child, and Child-only) across two initialization strategies (Adult PhonoQ and scratch). Generalization is evaluated against manual child-
speech annotations. On 1,352 consonant targets from 58 typically developing children, child-
speech adaptation improves voicing recognition across all supervision conditions, from 0.922 macro-F1 for Adult PhonoQ to 0.972–0.987 after adaptation. Manner is more sensitive to
alignment supervision: Adult+Child MFA reaches 0.804 and 0.796, compared to approximately 0.70 under Adult MFA supervision. Place remains comparatively strong across systems (0.871–0.902), although per-class performance varies substantially. The velar-fronting contrast is preserved across all seven model variants. Longitudinal UltraPhonix analysis further reveals speaker-specific velar and post-alveolar changes that are largely preserved across models and broadly consistent with reported clinical progress.
Read the full abstract ▾
All authors Abner Hernandez, Tomás Arias Vergara, Andreas Maier, Paula Andrea Pérez-Toro
Computer Vision
SmoothOperator: Enhancing Representations for Fine-grained Open-set Recognition via Modulated Label Smoothing
Thiru Thillai Nadarasar Bahavan, Yu Xia, Sachith Seneviratne et al. (+1 more)
Open Set Recognition (OSR) aims to enable models to accurately classify known classes while rejecting samples from unseen classes. A key challenge in OSR lies in the inability to model the unbounded distribution of unknown classes during training, often leading to the misclassification of samples from these classes. Rather than modeling unknowns, recent work shapes the feature space so that known classes are compact and well separated, and spherical representation learning methods have achieved strong results this way. Label smoothing has been identified as one of the key drivers of this success, yet it applies the same coefficient to every training sample, regardless of how well each sample is already embedded. We show that the spherical representation learning objectives used in OSR share a single
alignment–uniformity structure in which labels enter only through the
alignment term. Label smoothing therefore acts as an
alignment dial, and a fixed coefficient sets this dial to the same value for every sample. We propose a plug-in, SmoothOperator (SmoothOP), which sets the smoothing coefficient of each sample from its prominence, an
embedding-space signal measuring how clearly the sample's own class stands out against its strongest competing class. Our method integrates into four existing spherical representation learning methods at minimal training overhead. SmoothOP assigns strong smoothing to samples with high prominence, which reduces their
alignment and relaxes their pull. On the Semantic Shift
Benchmark, SmoothOP-augmented variants generally outperform their base objectives across
datasets, degrees of semantic shift, and OSR post-processors, with gains of up to 4.7% in AUROC, OSCR, and closed-set accuracy.
Read the full abstract ▾
All authors Thiru Thillai Nadarasar Bahavan, Yu Xia, Sachith Seneviratne, Saman Halgamuge
Cryptography and Security
AuraForge: Scaling Security Supervision for Training Coding Agents
Danqing Wang, Songwen Zhao, Harsh Sharma et al. (+4 more)
Coding
agents are now proficient enough to generate complex software applications from a single
prompt. As their capabilities have grown, human oversight has increasingly shifted from line-by-line code review toward hands-off evaluation of outcomes. However, recent studies have shown that such a transition exposes a critical risk: functional correctness alone does not guarantee a secure implementation. Despite growing
attention to code security, training safer coding
agents remains challenging because reliable security supervision is difficult to obtain at scale from real-world repositories. We introduce AuraForge to synthesize and validate executable security tests for training secure coding
agents. Our approach combines attack-oriented test synthesis,
language-extensible task construction, and safeguards against reward hacking. Using AuraForge, we construct AuraGym, a multi-
language and multi-CWE executable training gym: 679 executable feature-implementation tasks from 344 real-world repositories across Python, JavaScript, and TypeScript, covering 177 CWE categories. On the subset with human-written security tests, AuraForge produces about 3 times as many test cases on average and reduces the false-positive rate by 83.23%, allowing alternative secure implementations to receive correct supervision. Training Qwen3.5-4B with synthesized security tests gains larger improvements than human-written security tests (average 19.7 FuncPass and 6.2 SecPass vs. 14.9 FuncPass and 4.4 SecPass) on three
languages. These results demonstrate that AuraForge provides more diverse and reliable security supervision to train secure coding
agents.
Read the full abstract ▾
All authors Danqing Wang, Songwen Zhao, Harsh Sharma, Jierui Wang, Andre Vicente Duarte, Ivan Bercovich, Lei Li
Artificial Intelligence
Learning Multiple Timescales for Goal-Conditioned Reinforcement Learning
Pedro Robles Dutenhefner, Dikshant Shehmar, Wagner Meira Jr. et al. (+1 more)
Existing approaches to offline goal-conditioned reinforcement learning (GCRL) struggle with long-horizon tasks. Discounting shrinks value differences between distant states until they fall below the function approximation error, leaving the
agent with no signal for ranking states. Temporal abstraction, which treats k environment steps as a single transition, restores this signal at long range, but no single fixed k suits all state-goal distances: large k preserves value differences across long temporal distances while collapsing distinctions between nearby states, and small k does the reverse. We make this trade-off explicit and introduce Generalized Implicit Temporal Abstraction (GITA), which conditions a single value function on k. GITA trains one policy by aggregating advantage-weighted supervision across multiple k values, so scales assigning larger positive advantages to a state-goal pair contribute more strongly to its update. GITA does not need to choose between local resolution and long-range signal; it retains both without committing to a single k. On OGBench, GITA outperforms a broad range of offline GCRL baselines, raising average success rate across all tasks by 25 percentage points (73% relative improvement) over HIQL. It also improves over the strongest fixed-k method, OTA, by 7 percentage points (14% relative).
Read the full abstract ▾
All authors Pedro Robles Dutenhefner, Dikshant Shehmar, Wagner Meira Jr., Marlos C. Machado
Computer Vision
Geometric Similarity in VLM Low-Level Vision Representations
Shao-Jun Xia, Huixin Zhang, Zhen Lei et al. (+3 more)
Vision-
language models (VLMs) have emerged as powerful candidates for universal
vision backbones, with representative architectures including autoregressive (AR) models and
diffusion transformers (DiTs). Yet, adapting them efficiently for all-in-one low-level image restoration remains a challenge. Crucially, the field lacks an understanding of how VLMs organize hidden-layer representations and whether these structurally distinct paradigms share a common
geometric organization for pixel-level perception. Such shared organization is a prerequisite for building highly transferable, unified restoration VLMs and adapters. In this paper, we systematically investigate representational similarity across 24 low-level tasks spanning 5 categories. We propose GeoSim, a unified four-level framework that analyzes task-conditioned representations from global similarity, local geometry, sparse feature decomposition, and topological verification perspectives. Our formulation applies to the analysis of hidden states in AR models and feature maps in DiTs across same- and cross-task/model settings. Our results reveal the organizing principles of low-level visual representations while exposing their limits in cross-task and cross-model agreement. Ultimately, GeoSim provides an interpretability lens for probing latent transferability in low-level
vision and diagnosing model limitations in task- or model-specific scenarios.
Read the full abstract ▾
All authors Shao-Jun Xia, Huixin Zhang, Zhen Lei, Anlan Sun, Yuner Zhang, Xiaoyang Chen
Computational Complexity · Cross-listed from quant-ph
Beyond IP = PSPACE and QIP = PSPACE: Interactive Proofs in Arbitrary Physical Theories
Kishor Bharti
The equalities IP = PSPACE and QIP = PSPACE, the latter achievable with three messages, raise a basic question: how much of an interactive proof's power comes from the underlying physical theory? We study interactive proofs in general probabilistic theories, which include classical and
quantum theory. The answer depends on what the prover and verifier exchange and how the theory specifies efficient operations. When they exchange only classical messages, protocols in every theory satisfying our standard assumptions decide exactly PSPACE. For protocols with a
quantum verifier and
quantum messages, allowing a prover to use any theory containing
quantum theory does not increase the maximum acceptance probability. Thus, the three-message PSPACE result remains valid against such provers. When messages may be arbitrary systems, the interactive-proof class can strictly exceed PSPACE.
Read the full abstract ▾
Data Structures and Algorithms
Sparsification Framework for Directed Densest Subgraph
Slobodan Mitrović and Theodore Pan
We develop a new approach for computing approximate directed densest subgraphs (DDS). Our main result is a sparsification procedure that reduces a directed graph $G$ on $n$ vertices to a graph with $n \cdot \text{poly} \log n$ edges while preserving enough structure to recover an approximate DDS of $G$. Instantiating this framework in several
memory-constrained settings, we obtain the following improvements over the state of the art: In semi-streaming, we obtain a single-pass
algorithm that computes a $(1-\varepsilon)$-approximate DDS. Previously, the only semi-streaming
algorithm that computed a constant approximation of DDS was by Bahmani, Kumar, and Vassilvitskii (2012), providing a $0.5-\varepsilon$ approximation in $O(\log n)$ passes. Hence, our work completely closes the approximation gap between undirected and directed DS in the semi-streaming setting, matching the $(1-\varepsilon)$-approximate undirected DS
algorithm by Esfandiari, Hajiaghayi, and Woodruff (2016). In the near-linear-
memory MPC regime, we obtain an $O(1)$-round
algorithm for $(1-\varepsilon)$-approximate DDS, improving over the $O(\sqrt{\log n})$-round $(0.5-\varepsilon)$-approximation
algorithm of Mitrović and Pan (2024). In the sublinear-time setting, we obtain an
algorithm using $\tilde{O}(n)$ time, space, and oracle queries to compute a $(1-\varepsilon)$-approximate DDS, improving over the $\tilde{O}(n^{1.5})$ time, space, and query
algorithm of Esfandiari, Hajiaghayi, and Woodruff (2016).
Read the full abstract ▾
Performance · Cross-listed from physics.plasm-ph
Is Your AI Fast Enough to Run a Fusion Reactor?
Nathaniel Chen, Andrew Rothstein, Ricardo Shousha et al. (+4 more)
Machine learning models are increasingly used in feedback control loops for
nuclear fusion, where inference speed and predictable timing are critical. We summarize lessons from models deployed for control on the DIII-D tokamak and develop a
benchmark to compare inference backends across ten neural networks and model components from fusion control and diagnostic pipelines. For models greater than five million parameters, the CPU backends take tens to thousands of milliseconds, while GPU inference is substantially faster, suggesting an upper limit on CPU-oriented development for control. These results show why the deployment backend must be selected together with the model and its control-cycle budget.
Read the full abstract ▾
All authors Nathaniel Chen, Andrew Rothstein, Ricardo Shousha, Hiro Farre-Kaga, Peter Steiner, Azarakhsh Jalalvand, Egemen Kolemen