Machine Learning · Cross-listed from q-fin.RM
SHARC: SHAP-Based Interpretability in Machine Learning Risk Models for Regulatory Capital under ICAAP and CCAR
Ujjwala Vadrevu
The adoption of non-parametric machine learning models for regulatory capital estimation introduces a fundamental governance challenge: the inability to explain model outputs in a manner auditable by supervisory bodies. This 'black box' problem remains a major barrier to the adoption of Gaussian Process Regression (GPR) and related ML architectures in ICAAP and CCAR workflows despite their predictive advantages over traditional parametric approaches. This paper addresses this barrier through SHARC (SHAP for Regulatory Capital), an explainability framework for the Hybrid GPR-HS architecture and its stress-testing extension. SHapley Additive exPlanations (SHAP), derived from cooperative game theory and satisfying the properties of Local Accuracy, Missingness, Consistency, and Efficiency, are applied to Stressed Value-at-Risk (SVaR) outputs under three macro scenarios: West Asia War,
Climate Risk, and AI Bubble/Regulatory Burden. SHARC decomposes SVaR into baseline, mean-driven, and
volatility-driven components, enabling transparent linkage between scenario design and capital outcomes. Two findings emerge. First, SHARC consistently links non-linear SVaR outputs to underlying scenario inputs, confirming framework fidelity and providing auditable traceability of capital drivers. Second, under stress conditions, the mean return component (directional loss magnitude) dominates the variance component (
volatility baseline) in determining capital levels, with implications for capital limit-setting, position management, and hedging strategy. The results establish SHARC as a regulator-aligned explainability layer that makes the Hybrid GPR-HS framework fully auditable and consistent with FRTB, ICAAP Pillar 2, and CCAR transparency requirements.
Read the full abstract ▾
Machine Learning
PatchOptic for Shared-State LLM Workflows with Projected Views and Verified Structured Updates
Zhaoyu Bai and Jiaqi Cai
Agentic workflows often operate over shared, structured state. Because LLM context windows are limited, each model invocation is typically shown only the state fragment needed for the current workflow step, a pattern commonly known as progressive disclosure. Modern systems construct such model-facing views using grep-like keyword search,
retrieval-augmented generation (RAG), abstract-syntax-tree (AST) queries, and task-specific
agent skills. These methods make the read side manageable, but they do not define when a locally proposed rewrite is valid after it is applied back to the full state. The missing piece is a contract between local updates and global validity. We introduce PatchOptic, an optic-inspired interface for shared-state LLM workflows. Optics are compositional bidirectional accessors that describe how views of structured data are read and updated. PatchOptic borrows this view/update intuition and realizes it through projected reads and verified structured patches. Each workflow step declares a projected read view, an authorized write region, and a patch-source region. Beyond runtime enforcement, the same declaration yields a path-level footprint that supports delegation, sub-workflow composition, and static certificates for reordering independent steps within the same phase. We evaluate this design with PatchBench, a
benchmark with 46 cases across domains. The results show that projected reads reduce reported leakage and
token cost while preserving accepted-output quality under the strong actor. Runtime verification blocks declared workflow-contract violations before commit, and patch-read enforcement rejects compromised patch artifacts that use hidden sources.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CR
Full-range Binary Classifier Calibration for Stable Model Updates in Production
Konstantin Berlin
Detection models running in
adversarial environments face a malicious distribution that drifts rapidly while the benign distribution stays comparatively stable, so teams retrain and redeploy constantly to stay ahead of new threats. Retraining tends to change the output prediction scores, which breaks downstream users of the model. For these security-oriented models we need consistent false-positive rate (FPR) across all output values, whereas standard probability-calibration methods target class probability rather than an FPR contract. We introduce a method built on top of existing calibration primitives that targets the whole FPR curve, giving scores a consistent FPR meaning across deployments. On one held-out split, the observed relative FPR error was at most 2.3% from 10% down to 0.1% FPR and 7.2% at 0.01% FPR. The shipped artifact remains under 200 KB in measurements across calibration sets from 1K to 10M benign samples.
Read the full abstract ▾
Machine Learning
InvWeaver: Deductive Feedback for Invariant Synthesis in Interacting-Loop Programs
Guangyuan Wu, Weining Cao, Zehui Tan et al. (+4 more)
Loop invariant inference is a fundamental yet challenging problem in program verification. Recent LLM-aided guess-and-check techniques have shown strong performance on single-loop programs, but they often struggle with programs containing multiple interacting loops. This paper presents InvWeaver, a neuro-symbolic framework for synthesizing invariants for such programs. The key idea is to expose inter-loop dependencies and propagate proof obligations through a combination of loop-level abstraction, obligation-guided inference, and weakest-precondition-based refinement. We evaluate InvWeaver on a comprehensive
benchmark suite, including a newly curated
dataset derived from classic
algorithms. Experimental results show that InvWeaver substantially outperforms existing invariant inference methods, solving 72 out of 82 multi-loop
benchmark problems and maintaining strong performance on single-loop tasks.
Read the full abstract ▾
All authors Guangyuan Wu, Weining Cao, Zehui Tan, Yuan Yao, Hengfeng Wei, Taolue Chen, Xiaoxing Ma
Machine Learning
Parameter-Free Encoders Remain Viable for RDB Foundation Models
Linjie Xu and David Wipf
Given a relational database (RDB) storing heterogeneous tabular information, how can we predict missing (or future) values in some target column of interest? As the space of potential targets is vast across enterprise settings, it is preferable to avoid learning a new model from scratch each time there is a new prediction task. Frozen foundation models based on RDB-specific encoders provide a viable solution, but ideal design remains an open question. On the one hand, it has recently been argued that certain parameter-free subgraph encoders combined with single-table foundation models can achieve near SOTA performance, with no RDB-specific pre-training required. Meanwhile, other contemporary studies advocate for parameterized encoders pre-trained to exploit observable labels for learning task-specific representations. To address this ambiguity, we analyze RDB encoder properties specifically when labels are present as inputs, proving limitations on the potential efficacy of trainable encoder parameters. As empirical validation, we demonstrate that considerably simpler parameter-free encoders are still capable of strong performance across many relevant benchmarking tasks.
Read the full abstract ▾
Machine Learning
Breaking Structural Isolation: Scalable Graph Clustering via Community-Aware Sampling and Structural Entropy
Jingyun Zhang, Hao Peng, Jianxin Li et al. (+2 more)
Unsupervised graph clustering is a fundamental technique for uncovering underlying semantic patterns in large-scale networks. Although Graph Contrastive Learning has demonstrated promising performance, existing methods often suffer from the "structural isolation" issue during mini-batch training, making it challenging to capture cohesive community structures that characterize the global topological distribution. To address these challenges, we propose SCISE, a Scalable unsupervised graph Clustering framework that preserves structural Integrity by synergizing community-aware sampling with constrained Structural
Entropy. Specifically, we first introduce the Structural
Entropy Community Constraint operator (SECC), which optimizes structural information within a constrained solution space to mitigate community fragmentation and enhance partition cohesion. Second, to prevent global information loss during batch training, we design a Community-Aware Sampling Expansion (CSampE) mechanism that incorporates the community context of target nodes into sampling batches, effectively breaking structural barriers and preserving topological integrity. Finally, we devise a Structural Contrastive Learning (StructCL) module that refines edge weights based on intra-batch structural similarity, guiding the encoder to learn representations in a higher-order structural space. Extensive experiments on six mainstream
benchmark datasets demonstrate that SCISE significantly outperforms state-of-the-art
algorithms, with ablation studies and robustness analyses further validating its effectiveness and reliability for real-world large-scale graphs.
Read the full abstract ▾
All authors Jingyun Zhang, Hao Peng, Jianxin Li, Angsheng Li, Philip S. Yu
Machine Learning · Cross-listed from cs.CV
A Task-Driven Evaluation of UAV Detection and Tracking under Synthetic Fog
Amir Pouladi, Vesal Ahsani, Haijun Li et al. (+2 more)
Fog severely degrades the visibility of small unmanned aerial vehicles (
UAVs) in skydominant, long-range imagery, reducing the reliability of downstream detection and tracking. This paper presents a task-driven evaluation framework that links depth-aware synthetic fog generation, image restoration, object detection, and tracking within a unified pipeline. Given the practical difficulty of collecting and annotating foggy
UAV scenes, synthetic fog is generated from real clear-weather outdoor images containing
UAV targets using
monocular depth estimation and the atmospheric scattering model. Representative restoration methods from classical, convolutional neural network (CNN)-based, and
transformer-based families are first compared, after which the selected restoration model is integrated into the downstream perception pipeline. Detection is evaluated under both clean-only and fog-inclusive training regimes using multiple detector variants, while tracking-by-detection is assessed on clean, foggy, and restored
video sequences. Beyond image-level restoration metrics, the study evaluates how fog and restoration affect detection robustness and tracking performance. The results show that fog substantially degrades both detection and tracking, primarily through increased missed detections. Fog-inclusive training provides the most consistent improvement in robustness, whereas test-time restoration is most beneficial when the detector has been trained only on clean imagery. These findings show that restoration quality does not necessarily translate into proportional gains in downstream perception and therefore should be evaluated jointly with detection and tracking performance.
Read the full abstract ▾
All authors Amir Pouladi, Vesal Ahsani, Haijun Li, Homayoun Najjaran, Afzal Suleman
Machine Learning
Learnable Weighting of Intra-Attribute Distances for Categorical Data Clustering with Nominal and Ordinal Attributes
Yiqun Zhang and Yiu-ming Cheung
The success of categorical data clustering generally much relies on the distance metric that measures the dissimilarity degree between two objects. However, most of the existing clustering methods treat the two categorical subtypes, i.e. nominal and ordinal attributes, in the same way when calculating the dissimilarity without considering the relative order information of the ordinal values. Moreover, there would exist interdependence among the nominal and ordinal attributes, which is worth exploring for indicating the dissimilarity. This paper will therefore study the intrinsic difference and connection of nominal and ordinal attribute values from a perspective akin to the graph. Accordingly, we propose a novel distance metric to measure the intra-attribute distances of nominal and ordinal attributes in a unified way, meanwhile preserving the order relationship among ordinal values. Subsequently, we propose a new clustering
algorithm to make the learning of intra-attribute distance weights and partitions of data objects into a single learning paradigm rather than two separate steps, whereby circumventing a suboptimal solution. Experiments show the efficacy of the proposed
algorithm in comparison with the existing counterparts.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.AI
Governable Individuals: An Identity Layer for Embodied Agents That Keep Learning
Xue Qin, Simin Luan, Cong Yang et al. (+1 more)
Embodied artificial intelligence is moving from deployable models to persistent
agents that learn in the field, acquire skills and migrate across bodies. Governing such a system means governing an individual, not a model, and existing proposals (
agent identifiers, activity logs, guardrails) do not survive an
agent that keeps rewriting itself. We propose the governable individual: an
agent whose competence may change without bound, but whose authority,
memory schema, embodiment rights and capability roster can widen only through signed lifecycle transitions that update a public identity commitment. In our tests, neither learned judgement nor behavioural testing was sufficient to carry this on its own; the load-bearing layer must be architectural. We describe the abstraction, a runtime mechanism that realizes it, and the open problems in between.
Read the full abstract ▾
All authors Xue Qin, Simin Luan, Cong Yang, Zhijun Li
Machine Learning
AdaStop: Cost-Aware Early Stopping for DNN Test Selection
Bonan Shen, Wei-Jung Huang, Xin Liu et al. (+2 more)
Existing methods for testing deep neural networks (DNNs) primarily prioritize test inputs likely to reveal model faults under a fixed labeling budget. In practice, choosing that budget is difficult: too little testing misses failures, while too much incurs unnecessary labeling costs. This work studies the stopping problem in DNN testing. We formulate testing as a cost–benefit decision process in which labeling an input incurs cost $c$ and discovering a fault yields value $v$. Based on this formulation, we introduce AdaStop, a framework that estimates the marginal fault discovery rate during testing and stops labeling when the estimated rate falls below the threshold $\tau = c/v$. Experiments across multiple
datasets, architectures, and selection strategies show that $65$–$84\%$ of faults can be discovered using only $9$–$31\%$ of the labeling budget.
Read the full abstract ▾
All authors Bonan Shen, Wei-Jung Huang, Xin Liu, Jiazhou Gao, Tao Ning
Machine Learning · Cross-listed from cs.CV
From Fixed to Free Cameras: Calibration-Free View-Robust Vision-Language-Action Model
Wenhao Li, Xueying Jiang, Quanhao Qian et al. (+4 more)
Real-world
robot deployment rarely maintains the training-stage camera setup, where cameras often experience repositioning or remounting depending on actual scenarios. Existing view-robust
Vision-
Language-Action (VLA) policies tolerate such camera variations only when the camera extrinsics are explicitly provided, making them fragile and hard to use especially when view robustness is critical. We argue that the policy should not be told where the camera is, but rather figure it out by itself. To this end, we introduce Camera-Centric VLA (CamVLA), a new VLA model that decouples manipulation controls from camera geometry by predicting (i) a camera-centric end-effector action expressed in the local camera frame, and (ii) a 6-DoF hand-eye
matrix relating cameras to the
robot base. A deterministic
geometric transformation composes the two predictions into a
robot base-frame action. This disentangles how I should move in pose-independent camera-centric action generation from where I am looking from in camera-perspective
geometric grounding. The resulting policy is calibration-free, depth-free, and single-view, requiring only a single
monocular RGB image as the visual observation and task instruction at deployment. Evaluations in both simulation and real-world
robot data show that CamVLA consistently improves success rates across diverse unseen viewpoints. Project page: https://alibaba-damo-academy.github.io/CamVLA/.
Read the full abstract ▾
All authors Wenhao Li, Xueying Jiang, Quanhao Qian, Deli Zhao, Shijian Lu, Gongjie Zhang, Ran Xu
Machine Learning
Weak-to-Strong Generalization via Direct On-Policy Distillation
Shiyuan Feng, Huan-ang Gao, Haohan Chi et al. (+7 more)
Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving
language-model
reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollouts during training. As models scale, post-training itself becomes a bottleneck. We study a weak-to-strong alternative: run RL on a smaller model where rollouts are cheaper, then reuse what that RL run learned to improve a stronger target model. Directly distilling the post-RL weak teacher is not enough, because the teacher's final policy mixes useful RL gains with the limitations of the smaller model. We propose Direct On-Policy
Distillation (Direct-OPD), which transfers the teacher's RL-induced policy shift instead. Direct-OPD compares the post-RL teacher with its own pre-RL reference and treats their log-ratio as a dense implicit reward for the student. In plain terms, the checkpoint pair tells us which actions RL made the weak model more or less likely to take, and Direct-OPD applies that signal on the stronger student's own on-policy states. This directly reuses the weak model's RL supervision signal without running sparse-reward RL on the target model. Empirically, Direct-OPD consistently leverages weaker teachers to improve stronger target models; notably, it boosts Qwen3-1.7B from 48.3% to 58.3% on AIME 2024 in just 4 hours on 8 A100 GPUs. It outperforms step-matched direct RL and enables the sequential composition of multiple policy shifts. Our results show that RL outcomes can be reused across model scales as implicit reward signals, not merely as final models to imitate.
Read the full abstract ▾
All authors Shiyuan Feng, Huan-ang Gao, Haohan Chi, Hanlin Wu, Zhilong Zhang, Zheng Jiang, Bingxiang He, Wei-Ying Ma, Ya-Qin Zhang, Hao Zhou
Machine Learning · Cross-listed from astro-ph.IM
Interpretable Human-Label-Free Deep Learning for Real-Bogus Classification with Uncertainty Quantification
Raphaël Bonnet-Guerrini, Bruno Sanchez, Dominique Fouchez et al. (+5 more)
Time-domain surveys generate many transient candidates, making Real-Bogus classification a critical step in automated discovery pipelines. Reliable labels are costly, while community labels can be noisy and survey-dependent. We aim to develop a Real-Bogus classification framework that can be trained without human-labeled data using injected transients and bogus-dominated survey data, remains robust under strong class contamination, and provides calibrated uncertainty quantification. We combine simulated transient injections with a contaminated survey class and train a dual-network model using asymmetric co-teaching for classes with different label-noise levels. We evaluate performance on a
benchmark subset and analyze the learned representation with latent-space visualization tools. For uncertainty quantification (UQ), we compare MC dropout and deep ensembles and propose a low-cost hybrid strategy that exploits the dual-network setting to improve calibration. We extend the evaluation to the light-curve domain to assess recovery of light-curve classes. The method achieves strong Real-Bogus performance on the labeled subset and remains stable under severe class contamination. It recovers transient light-curve classes with high fidelity, while single-source identification is limited by ambiguity in light-curve-derived labels. Our hybrid UQ approach achieves competitive calibration relative to more expensive ensemble baselines. Latent-space analyses indicate that uncertainty aligns with the decision boundary and reveal subclasses within the bogus population. Our results show that injection-driven, weakly supervised training can enable scalable and consistent Real-Bogus classification without human-labeled training data while providing calibrated uncertainties. The method is suited for transfer to forthcoming surveys by re-running the injection-based training pipeline.
Read the full abstract ▾
All authors Raphaël Bonnet-Guerrini, Bruno Sanchez, Dominique Fouchez, Benjamin Racine, Maya Guy, Mariam Sabalbal, Manal Yassine, Vincenzo Piuri
Machine Learning · Cross-listed from cs.AI
LLM-as-a-Verifier: A General-Purpose Verification Framework
Jacky Kwok, Shulu Li, Pranav Atreya et al. (+6 more)
Scaling pre-training, post-training, and test-time compute have become the central paradigms for improving the capabilities of LLMs. In this work, we identify verification, the ability to determine the correctness of a solution, as a new scaling axis. To unlock this and demonstrate its effectiveness, we introduce LLM-as-a-Verifier, a general-purpose verification framework that provides fine-grained feedback for agentic tasks without requiring additional training. Unlike standard LM judges that
prompt LLMs to produce discrete scores for candidate solutions, LLM-as-a-Verifier computes the expectation over the distribution of scoring
token logits to generate continuous scores. This probabilistic formulation enables verification to scale along multiple dimensions: (1) score granularity, (2) repeated evaluation, and (3) criteria decomposition. In particular, we show that scaling the scoring granularity leads to better separation between positive and negative solutions, resulting in more calibrated comparisons. Moreover, scaling repeated evaluation and criteria decomposition consistently lead to additional gains in verification accuracy through variance and complexity reduction. We further introduce a cost-efficient ranking
algorithm for selecting the best solution among candidates using the verifier's continuous scores. LLM-as-a-Verifier achieves state-of-the-art performance on Terminal-Bench V2 (86.5%), SWE-Bench Verified (78.2%), RoboRewardBench (87.4%), and MedAgentBench (73.3%). Beyond verification, the fine-grained signals from LLM-as-a-Verifier can also serve as a proxy for estimating task progress. We build an extension for Claude Code, enabling developers to monitor and improve their own agentic systems. Finally, we show that LLM-as-a-Verifier can provide dense feedback for RL, improving the sample efficiency of SAC and GRPO on robotics and mathematical
reasoning benchmarks.
Read the full abstract ▾
All authors Jacky Kwok, Shulu Li, Pranav Atreya, Yuejiang Liu, Yixing Jiang, Chelsea Finn, Marco Pavone, Ion Stoica, Azalia Mirhoseini
Machine Learning
What Does a Discrete Diffusion Model Learn? One Reverse Process in Three Coordinates
Rodrigo Casado Noguerales, Bernhard Schölkopf, Thomas Hofmann et al. (+1 more)
What does a discrete
diffusion model actually learn? The literature trains denoisers, bridge plug-ins, and concrete scores, each deriving its own ELBO objective under different conventions. This fragmentation makes the object being learned unclear and ELBO values difficult to interpret and compare. All three methods learn the same object in different coordinates: the exact time-reversal of the noising process. We first prove the Oracle Distance identity: for any noising process, the negative ELBO is not just a bound on the likelihood; it equals the data
entropy plus the path KL divergence from the reversed noising process to the learned one. Next, for
token-factorizing noise, we show that denoiser, bridge plug-in, and concrete score are three coordinates of this reverse process, derive exact conversions between them, and identify the probability distribution learned by each. In particular, contrary to its standard interpretation, the bridge plug-in learns not a denoiser but the cavity distribution: the clean
token given the noisy sequence excluding its own position. We confirm the theory by training all three coordinates with masked, uniform, and GIDD kernels on OpenWebText under a shared recipe, and by training uniform-
diffusion solvers on four logic tasks. The learned coordinates agree after conversion, while confusing the cavity with the denoiser can severely degrade sampling quality. Together with exact boundary-term formulas, closed-form NELBO calibrations at initialization, and a clock interpretation of time schedules, these results provide a conceptual framework and a mechanical recipe for deriving, implementing, and comparing discrete
diffusion processes.
Read the full abstract ▾
All authors Rodrigo Casado Noguerales, Bernhard Schölkopf, Thomas Hofmann, Aran Raoufi
Machine Learning
TabPack: Efficient Hyperparameter Ensembles for Tabular Deep Learning
Yury Gorishniy, Akim Kotelnikov, Ivan Rubachev et al. (+1 more)
In deep learning for tabular data, efficient ensembles of multilayer perceptrons (MLPs) have recently emerged as effective and practical architectures. Existing methods of this kind use the same hyperparameters for all underlying MLPs, which requires hyperparameter tuning for achieving the best performance. In this work, we introduce TabPack, an efficient MLP ensemble with strong out-of-the-box performance and reduced reliance on traditional tuning. In a single run, TabPack samples and trains many MLPs with different hyperparameters efficiently in parallel and selects ensemble members on the fly during training. Thus, TabPack only requires specifying ranges from which to sample MLP hyperparameter rather than exact hyperparameter values, which naturally demands less precision for good performance. In experiments on medium-to-large public
datasets, TabPack with default settings performs on par with extensively tuned prior methods, thus substantially reducing effort and compute resources needed to achieve competitive results on tabular tasks. Notably, running the default TabPack configuration on a modern MacBook took less time than tuning some baselines on an industry-grade GPU.
Read the full abstract ▾
All authors Yury Gorishniy, Akim Kotelnikov, Ivan Rubachev, Artem Babenko
Machine Learning
CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents
Yujiang Li, Zhenyu Hou, Yi Jing et al. (+2 more)
Long-horizon agentic LLMs are increasingly limited by finite context windows, as extended interaction trajectories can exceed the maximum context length before a task is completed. Context compaction offers a natural solution by summarizing previous interaction states and continuing the rollout under a compressed context, but incorporating compaction into reinforcement learning remains underexplored. We propose CompactionRL, a reinforcement learning strategy to train long-horizon agentic LLMs with context compaction. Our approach jointly optimizes task execution and summary generation with
token-level loss normalization and cross-segment generalized advantage estimation. This design enables the LLM
agents to learn from compacted long-horizon trajectories. We train CompactionRL on top of open models and observe consistent performance gains on agentic coding tasks. CompactionRL enables the open GLM-4.5-Air model (106B-A12B) to achieve Pass@1 scores of 66.4% on SWE-bench Verified and 26.2% on Terminal-Bench 2.0, exceeding the base model under inference-time compaction by 6.6 and 4.9 points, respectively. Built upon GLM-4.7-Flash (30B-A3B), CompactionRL improves Pass@1 by 5.5 and 6.7 points against the base model, reaching 56.0% on SWE-bench Verified and 20.2% on Terminal-Bench 2.0. CompactionRL is thus deployed in the RL pipeline for training the open GLM-5.2 model (750B-A40B).
Read the full abstract ▾
All authors Yujiang Li, Zhenyu Hou, Yi Jing, Jie Tang, Yuxiao Dong
Machine Learning · Cross-listed from stat.ML
Fitted Occupancy-Ratio Evaluation without Bellman Completeness
Lars van der Laan and Nathan Kallus
Occupancy ratios correct distribution shift in offline reinforcement learning and are central to off-policy evaluation. Existing primal-dual and minimax methods typically estimate these ratios by enforcing occupancy-balance moments over a critic class. We propose fitted occupancy-ratio evaluation (FORE), a fitted fixed-point method that characterizes the discounted occupancy ratio through an adjoint Bellman recursion. At each iteration, FORE solves a single-level density-ratio objective on one-step-transition data, thereby projecting the adjoint Bellman image onto a log-ratio class in Kullback-Leibler (KL) divergence. Unlike analyses of fitted Q-evaluation, which typically require value-function realizability together with Bellman completeness or projected-operator stability, our central approximation condition is just realizability of the discounted occupancy ratio itself. Under this condition, the population KL-projected recursion contracts in relative
entropy toward the true ratio by virtue of the adjoint Bellman operator being a KL-contraction. For the empirical recursion, we establish finite-sample regret bounds that yield convergence in KL up to approximation error and a statistical error governed by the complexity of the ratio hypothesis class. When full coverage fails, we introduce coverage-stopped FORE, which targets the discounted occupancy accumulated before the first uncovered state-action pair and yields a conservative lower bound on target-policy value for nonnegative rewards. The fitted ratio supports direct value estimation by reward reweighting, occupancy-weighted fitted Q-evaluation, and doubly robust estimation that combines the fitted ratio with a fitted Q-function. Together, these results identify discounted occupancy-ratio realizability as a sufficient condition for offline policy evaluation without any completeness assumptions.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.RO
GaP: A Graph-as-Policy Multi-Agent Self-Learning Harness For Variational Automation Tasks
Kaiyuan Chen, Shuangyu Xie, Letian Fu et al. (+21 more)
For
robots to work reliably in commercial and industrial applications, recent advances in agentic coding systems have potential to combine interpretable
robot programming with the open-world adaptability of model-free policies. We focus on "Variational Automation" (VA), a class of tasks that have larger variations in object geometry and pose than fixed automation. Model-free policies often struggle to close the reliability gap for VA tasks, which must be executed persistently and reliably in commercial and industrial applications. Motivated by prior work on Task and Motion Planning (TAMP) and the
Robot Operating System (ROS), we introduce Graph-as-Policy (GaP), a multi-
agent coding harness that generates directed computation graphs with perception, planning, and control nodes from a Modular Open
Robot Skill Library (MORSL). GaP generates an internal simulation environment to rehearse task instances with different graphs in parallel to iteratively refine the graph structure and parameters to improve success rates and throughput. Evaluation with 8 new open VA task
benchmarks, 4 in simulation and 4 in real-world, suggests that GaP can achieve success rates that significantly outperform baselines. Details, code, and data: https://graph-
robots.github.io/gap
Read the full abstract ▾
All authors Kaiyuan Chen, Shuangyu Xie, Letian Fu, Justin Yu, William Pacini, Sandeep Bajamahal, Hudson Kim, Jaimyn Drake, Daehwa Kim, Haoru Xue, Jonathan Francis, Christian Juette, Peter Schaldenbrand, Muhammet Yunus Seker, Ruwan Wickramarachchi, Uksang Yoo, Guanzhi Wang, Adithyavairavan Murali, Balakumar Sundaralingam, S. Shankar Sastry, Yuke Zhu, Linxi "Jim" Fan, Spencer Huang, Ken Goldberg
Machine Learning · Cross-listed from cs.CL
Faithfulness to Refusal: A Causal Audit of Neuron Selectors
Ananth Eswar, Pratinav Seth, Utsav Avaiya et al. (+1 more)
Attribution scores increasingly identify which
neuron rows of a
language model matter for applications such as
pruning, interpretability, and editing for safety, yet whether they identify causally important rows is rarely tested directly. We address this with two paired audits built on one-shot
neuron-row zeroing. We first audit selectors at the
language-modeling level: attribution methods substantially outperform activation and magnitude-based baselines at identifying dispensable rows across five LLMs. We then adapt the same intervention into a behavior test by driving it with a contrastive harmful-versus-benign signal; the attributed rows are sufficient to install refusal on hate and crime while keeping benign over-refusal low and preserving
language model fluency, and specific in that layer-matched random controls at the same depths fail. Highly rank-stable selectors can be among the least causally valid. Refusal moreover lives in a redundant subspace, where different attribution methods install it through largely disjoint row sets, so the recovered edit is one realization of a sufficient set rather than a unique mechanism. Together, these findings show that rank-stability proxies miss the kinds of selector failures a direct causal audit can surface.%
Read the full abstract ▾
All authors Ananth Eswar, Pratinav Seth, Utsav Avaiya, Vinay Kumar Sankarapu
Machine Learning · Cross-listed from cs.CR
Selective Disclosure Watermarking for Large Language Models
Xuyang Chen, Xiang Li, Yangxinyu Xie et al. (+1 more)
Watermarking methods embed imperceptible and verifiable signals into text generated by large
language models (LLMs). Existing approaches include zero-bit schemes for distinguishing synthetic text from human writing and multi-bit schemes for
embedding metadata. However, current multi-bit watermarking methods do not allow selective disclosure: verifying any part of the watermark requires revealing the entire embedded message. This lack of control leads to unnecessary information exposure and raises
privacy concerns. We propose Hierarchical Vocabulary Routing (HeRo), a watermarking framework that enables selective disclosure of embedded metadata. The method recursively partitions the vocabulary and distributes watermark information across hierarchical layers, so that different verifiers can decode only the portions of the payload corresponding to their access level. We show that the proposed scheme preserves the unbiasedness of the underlying sampling process and thus maintains text quality. Experiments demonstrate that our framework supports fine-grained access control while achieving high detection accuracy and low latency. Code is available at https://github.com/xuyangc03/hero-watermark.
Read the full abstract ▾
All authors Xuyang Chen, Xiang Li, Yangxinyu Xie, Qi Long
Machine Learning · Cross-listed from cs.CV
Multiplayer Interactive World Models with Representation Autoencoders
Anthony Hu, Václav Volhejn, Adrien Ramanana Rahary et al. (+24 more)
We introduce the first multiplayer world model for highly dynamic environments governed by complex physical interactions. Whereas single-player world models treat the other
agents as part of the environment, ours conditions on the action streams of multiple
agents, learning to attribute changes in the scene to the correct player and to stay coherent under arbitrary combinations of their actions. We study this problem in the game of Rocket League, where players compete and cooperate under fast, tightly coupled dynamics. Trained on 10,000 hours of gameplay collected with publicly available bots, our 5-billion-parameter latent
diffusion model generates four-player matches in real time, producing 20 frames per second on a single Nvidia B200 GPU. Although trained only on short clips, its rollouts stay stable far beyond the training horizon: distributional quality holds steady out to five minutes, the longest horizon we measure, and in practice we observe rollouts continuing for hours with no sign of collapse. We systematically investigate the central design choices: the
video codec, the generative objective, and the multiplayer conditioning scheme. In addition, we characterize how behavior changes with model and data scale, including the capabilities that emerge and the failure modes that persist. We further develop targeted evaluations that probe the model's physical understanding rather than visual appearance alone. To support continued research on multiplayer world models, we release our
dataset, our full training and inference codebase, and a live demo.
Read the full abstract ▾
All authors Anthony Hu, Václav Volhejn, Adrien Ramanana Rahary, Chris Mulder, Aditya Makkar, Alyx Liao, Amélie Royer, Manu Orsini, Adam Jelley, Eloi Alonso, Florian Laurent, Fredrik Norén, James Swingos, Jan Hünermann, Kent Rollins, Lucas Hosseini, Matthieu Le Cauchois, Maxim Peter, Pim de Witte, Tim Brown, Vincent Micheli, Moritz Böhle, Gabriel de Marmiesse, Viktoriia Sharmanska, Lucia Specia, Michael Black, Patrick Pérez
Machine Learning
TREK: Distill to Explore, Reinforce to Refine
Yuanda Xu, Zhengze Zhou, Kayhan Behdin et al. (+10 more)
Group Relative Policy Optimization (GRPO) is effective when the current policy already samples useful
reasoning trajectories, but it stalls on hard
prompts whose correct solution modes lie outside the student's on-policy support. We propose TREK (Teacher-Routed Exploration via Forward KL), a simple staged procedure that uses
distillation not for imitation but for exploration support expansion. A key advantage of TREK is its generality: because it only consumes verified output trajectories, it can use an external black-box teacher, a white-box teacher, or the same model given additional inference-time context, and it can efficiently identify which hard-
prompt samples are most worth consolidating even when teacher internals are unavailable. TREK first identifies
prompts where the unaided student has very low pass rate, queries a proposal source to produce verified candidate solutions, keeps the top-$r$ proposals ranked by current student likelihood, applies a short forward-KL phase to pull those verified modes into the student's support, and then returns to standard on-policy GRPO refinement. On mathematical
reasoning, TREK with DeepSeek-V4 proposals improves Qwen3 models across all tested scales on AIME 2024 and AIME 2025; for Qwen3-8B, it improves AIME 2025 from 36.9 to 40.3 and AIME 2024 from 47.9 to 51.1 (avg@16), while the self-context variant reaches 38.5 and 49.6 without an external teacher. On agentic tasks, TREK raises ALFWorld success rate from 75.8 to 82.8 and ScienceWorld success rate from 12.5 to 26.7; notably, on the hardest task types, TREK achieves high success rates early in training while unaided GRPO requires substantially more optimization steps to reach comparable levels.
Read the full abstract ▾
All authors Yuanda Xu, Zhengze Zhou, Kayhan Behdin, Jelena Markovic-Voronov, Hejian Sang, Xiaomin Li, Wenhui Zhu, Xinchen Du, Aida Rahmattalabi, Ran He, Sen Na, Zhipeng Wang, Alborz Geramifard
Machine Learning
How Far is Too Far? Defining the Distance Threshold for Verification Siamese Networks
Heloísa Dias Viotto, Cauê Samonek, Lucas Garcia Pedroso et al. (+3 more)
Siamese verification networks are widely used to compare items such as faces, cars, or signatures. In these scenarios, the network is trained to learn an
embedding space in which similar objects are mapped closer together, while dissimilar objects are mapped further apart. Two objects are considered to belong to the same class (e.g., the same person in two different images) when the distance between their
embeddings falls below a predefined threshold. Defining this threshold, however, is a non-trivial task and typically requires labeled data. In this work, we assume that the distribution of distances produced by a siamese verification network can be approximated by a bimodal function. Based on this assumption, we propose an unsupervised method to determine the verification threshold by identifying the minimum point between the two modes. The proposed approach does not require annotated samples, enabling the verification threshold to be updated directly in the deployment environment without the cost of manual labeling. We evaluate our method on four
datasets: MNIST, CIFAR-10, LFW, and PKLot. The results indicate that the proposed approach achieves an average verification accuracy of 94%, comparable to the Equal Error Rate method, while eliminating the need for labeled data.
Read the full abstract ▾
All authors Heloísa Dias Viotto, Cauê Samonek, Lucas Garcia Pedroso, Marcos Sunye, André Abed Grégio, Paulo Lisboa de Almeida
Machine Learning · Cross-listed from cs.CV
Shape Over Intensity: Directional Topological Encoding for False Positive Reduction in Intracranial Aneurysm Detection
Akshay Gokhale and Mansi Dhamne
Automated detection of intracranial aneurysms (IAs) from CT angiography (CTA) is severely hindered by high false-positive rates. Convolutional neural networks (CNNs) rely on local pixel intensities, causing systematic confusion between saccular aneurysms and vascular bifurcations - a problem especially acute for small lesions (<3 mm), where detection sensitivity falls below 60%. We propose a plug-and-play,
topology-aware false-positive reduction framework evaluating the Smooth Euler Characteristic Transform (SECT) - a directional representation encoding global 3D vascular geometry independently of intensity - against persistence-based summaries (Persistence Images and Landscapes), tested on a stratified subset of the RSNA 2025
dataset. SECT achieves an AUC of 0.943, substantially outperforming direction-agnostic methods (AUC ~0.68), and exhibits a clinical performance inversion: it excels on the sub-3 mm cohort, maintaining 0.943 AUC and 78.5% sensitivity at 95% specificity. The representation is also scanner-agnostic, achieving 0.927 mean AUC under leave-one-scanner-out (LOGO) validation across four manufacturers. By capturing asymmetric
geometric invariants rather than intensity profiles, SECT reliably resolves the primary structural confounder in IA detection, positioning it as a robust downstream filter for hybrid deep-learning diagnostic pipelines.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CL
How Much is Left? LLMs Linearly Encode Their Remaining Output Length
Mohamed Amine Merzouk, Dmitri Carpov, Mirko Bronzi et al. (+2 more)
Large
language models generate one
token at a time, yet their responses show remarkably consistent length structure: step-by-step solutions converge in predictable
token counts,
retrievals stop after a few sentences, retractions extend responses by measurable amounts. We ask whether the model carries an internal estimate of how much response remains. Training minimal-capacity linear probes on frozen hidden states of three open-weight 7-8B models across seven completion-style
datasets, we find three converging pieces of evidence. First, total response length is linearly decodable from the
prompt's last hidden state alone, before any output is emitted. Second, probe directions trained on natural-
language datasets transfer broadly, including to controlled synthetic completions never seen in training, outperforming a statistical baseline; the converse direction generally fails, and this asymmetry is itself informative. Third, on curated high-loss completions, the probe's per-position estimate shifts upward at the moment the model retracts and restarts a partial solution, a directional behavior no position-only predictor can reproduce (qualitative, not aggregate). We frame this as approximate estimation of remaining generation length, distinct from exact-counting impossibility results for
transformers, and interpret it as evidence that LLMs maintain a plan-like internal representation of output length (decodable, not necessarily used causally).
Read the full abstract ▾
All authors Mohamed Amine Merzouk, Dmitri Carpov, Mirko Bronzi, Damiano Fornasiere, Adam Oberman
Machine Learning · Cross-listed from quant-ph
Quantum Spectral Anomaly Detection
Yewei Yuan, Michele Minervini, Mark M. Wilde et al. (+1 more)
A core task in
quantum anomaly detection is to compute an anomaly score that quantifies how strongly a test
quantum state deviates from a given
quantum dataset assumed to be normal. Classically, principal component analysis (PCA) for centered data computes the anomaly score by evaluating the test sample relative to the subspace spanned by the selected leading eigenvectors. However, for
quantum data that lack a standard centering, explicitly recovering principal eigenvectors, constructing full Gram
matrices, or loading
quantum-random-access-
memory-style data can be more costly than estimating the anomaly score itself. To avoid these costs, we propose
Quantum Spectral Anomaly Detection (QSPADE), which computes PCA-like anomaly scores directly from the spectrum of the average state of the normal
dataset. By replacing hard PCA rank selection with a smooth, temperature-controlled spectral threshold, QSPADE makes near-threshold spectral components contribute partially to the anomaly score. This makes the score vary continuously rather than jump when a borderline component is included or excluded, and makes it less sensitive to noise or arbitrary hard cutoffs near the threshold. In the zero-temperature limit, QSPADE recovers the hard-projector PCA score. The proposed measurement-based
quantum detector can be calibrated with a sample complexity independent of the data dimension. Numerical simulations show that QSPADE behaves like kernel-PCA on encoded classical data and detects changes across a transverse-field Ising transition without predefined order parameters. Consequently, QSPADE gives an efficient framework for both
quantum-kernel anomaly detection on encoded classical data and the monitoring of
quantum-native systems where diagnostic observables are unknown.
Read the full abstract ▾
All authors Yewei Yuan, Michele Minervini, Mark M. Wilde, Nana Liu
Machine Learning
Biologically Informed Deep Neural Networks for Multi-Omic Integration, Pathway Activity Inference and Risk Stratification in Cancer
Pedro Henrique da Costa Avelar, Le Ou-Yang, Min Wu et al. (+1 more)
Integrating complex, multi-omics data presents significant challenges. Existing approaches often face a trade-off between model interpretability and representational capacity, with most either relying on post-hoc interpretation or use linear models that may overlook complex interactions. We report Pathway Activity Autoencoders for the multi-omics setting, which embed prior knowledge via pathway-informed architectural constraints, fostering interpretability, while preserving representational power. Our multi-omic framework is applied in the context of breast cancer and is evaluated in survival prediction and subtype classification with results indicating a positive effect of integration. We conduct analysis of individual omics layer impact on end-task performance, revealing that gene,
protein, and microRNA expression layers provide the strongest contribution. Repeatability studies indicate that, while dropout improves model robustness and consistency, excessive regularisation can reduce predictive performance. Finally, visualizations of the learned feature space illustrate the framework's intrinsic transparency and clinical relevance. The results underscore the value of multi-omic integration and delineate the impact of individual omics layers, establishing practical guidelines for integration within our framework. Overall, our pathway activity autoencoder frameworks yield superior latent representations that are biologically meaningful and are directly translatable into clinically relevant insights.
Read the full abstract ▾
All authors Pedro Henrique da Costa Avelar, Le Ou-Yang, Min Wu, Sophia Tsoka
Machine Learning
Learning Only What Valid Adapters Can Express: Subspace-Constrained Adaptation Against Fine-Tuning Poisoning
Fabien Polly
Parameter-efficient fine-tuning still leaves a broad space of behavior-changing updates reachable, so a poisoned objective can be represented and optimized. We study an alternative: adaptation constrained to the subspace estimated from a trusted pool of existing task adapters. On flan-t5-large with 196 public LoRA adapters, we show that (1) the functionally relevant content of an adapter lies in a low-dimensional shared subspace, 30 to 38 percent of its weight norm being redundant under the evaluated task distributions; (2) gradient adaptation restricted to 128 coordinates on this subspace matches full LoRA fine-tuning on clean classification data, while under targeted label inversion LoRA collapses to 3-26 percent exact match and the constrained learner keeps 62-96 percent on the tasks the pool covers; (3) the constrained learner cannot fit corrupted data, its adaptation loss separating clean from garbage by two orders of magnitude (120x), an out-of-distribution signal without an extra detector; and (4) against an adaptive
backdoor attacker who optimizes within the subspace, the attack is blocked (8 percent success versus 100 for LoRA) on the task where its target behavior is unlike anything in the pool, and only partially blocked (85 percent) when the target coincides with a common pool behavior. On these two tasks the outcome is consistent with how close the target is to the pool's directions, which suggests but does not establish a pool-relative boundary. The mechanism trades peak plasticity for these properties: on tasks the pool covers poorly, unconstrained fine-tuning wins, and the protection assumes the pool itself is trusted. Code and data are public.
Read the full abstract ▾
Machine Learning
Air Quality Downscaling with Station-Guided Pseudo-Supervision
Guorun Wang, Simone Foti, Andreas D. Demou et al. (+5 more)
Super-resolving coarse atmospheric fields to local PM$_{2.5}$ variations is uniquely challenged by a mismatch in spatial support: while pixels represent regional averages, ground-truth observations are discrete, unaligned samples of a continuous spatial signal. To bridge this gap, we present a station-guided framework for high-resolution PM$_{2.5}$ downscaling over Europe. Taking coarse CAMS atmospheric composition fields alongside heterogeneous side information (i.e., human activity, land cover, elevation, satellite aerosol observations, and wind fields) our framework jointly super-resolves ($\times 40$, $\approx$ 1 km) and bias-corrects CAMS rasters, without relying on temporal sequence modelling. To address the challenge of densely supervising our multi-scale
transformer network with sparse in-situ data, we introduce a time-agnostic propagation strategy that utilises spatial Gaussian blending of interpolated OpenAQ observations. Extensive qualitative and station-level evaluations across Europe demonstrate that our model recovers fine-grained spatial structures and effectively mitigates localised CAMS biases.
Read the full abstract ▾
All authors Guorun Wang, Simone Foti, Andreas D. Demou, Leonidas Kotoulas, Theodoros Christoudias, Alexandros Koliousis, Mihalis Nicolaou, Stefanos Zafeiriou