Machine Learning · Cross-listed from cs.SE
A semantic mutation metric for metamorphic relation adequacy in scientific computing programs
Meng Li, Xiaohua Yang, Jie Liu et al. (+1 more)
Context. Metamorphic Testing addresses the test-oracle problem in scientific computing, but classical Mutation Score operates on syntactic AST mutations and misses domain semantics. Objective. We propose the Semantic Mutation Score (SMS), built on five domain-semantic operators (Conservation Erosion, Operator Substitution, Hyperparameter, Trajectory Flip, Structural Injection). SMS degenerates almost everywhere to MS in a characterised limit, so any SMS-based conclusion remains consistent with prior mutation-testing literature in the classical regime. Method. A 12-PUT x 5-MP design over four single-output float-to-float classes (numeric, probabilistic, surrogate, machine-learning) is paired with a three-layer attribution classifier separating true semantic faults from tolerance, OOD, statistical, and artefact categories. A same-source / cross-source ablation under an identical
prompt isolates the LLM-source-diversity contribution. LLM-generated mutants are compared against a default-configuration
cosmic-ray syntactic pool at the AST-normalised level. Results. The pre-registered large-effect threshold for Cliff's delta is not met under the point-estimate criterion; the observed effect lies in the medium-effect range. Cross-source pooling under an identical
prompt does not appreciably shift delta, indicating that LLM identity is not the lever within this design. AST-level overlap between LLM-generated and default
cosmic-ray syntactic mutants is small; the Hyperparameter, Structural Injection, and Trajectory Flip classes are unreachable under default first-order syntactic configurations. Conclusion. SMS is a backward-compatible adequacy metric for domain-semantic metamorphic-relation sets in scientific computing. The first-order unreachability evidence is independent of the effect-size question.
Read the full abstract ▾
All authors Meng Li, Xiaohua Yang, Jie Liu, Shiyu Yan
Machine Learning
DP-SelFT: Differentially Private Selective Fine-Tuning for Large Language Models
Haichao Sha, Zihao Wang, Yuncheng Wu et al. (+2 more)
Large
language models (LLMs) are commonly adapted to downstream tasks through fine-tuning, but fine-tuning data often contains sensitive information that may be leaked by the resulting model. Differential
privacy (DP) offers formal protection against such leakage, yet DP fine-tuning of LLMs still suffers from substantial utility degradation due to gradient clipping and noise injection. Existing work improves this trade-off by combining DP with parameter-efficient fine-tuning methods such as LoRA, which constrain the form of updates. In this work, we study a complementary direction: selective fine-tuning, which constrains where updates are applied. We propose DP-SelFT, a framework for differentially private selective fine-tuning of LLMs. DP-SelFT addresses three DP-specific challenges in parameter selection: avoiding repeated
privacy cost, improving stability under noisy estimates, and selecting parameters that remain useful under clipped and noisy updates. It first constructs a lightweight DP synthetic
dataset and performs selection only on this synthetic data, so the selection stage incurs no additional
privacy cost. It then conducts layer-level selection by temporarily training candidate layer subsets on a synthetic training split and evaluating them on a synthetic validation split. Crucially, this temporary training is performed under a perturbation regime matched to downstream DP fine-tuning, with worst-case perturbations of the same scale as DP noise. This favors layer subsets that are not only learnable but also robust to noisy private updates. Experiments on
benchmark tasks show that DP-SelFT consistently improves the
privacy–utility trade-off over existing DP fine-tuning baselines under the same
privacy guarantees.
Read the full abstract ▾
All authors Haichao Sha, Zihao Wang, Yuncheng Wu, Hong Chen, Wei Dong
Machine Learning
MATE: Solving Contextual Markov Decision Processes with Memory of Accumulated Transition Embeddings
Himchan Hwang, Hyeokju Jeong, Gene Chung et al. (+3 more)
We propose MATE, a simple yet effective
memory architecture for solving Contextual Markov Decision Processes (CMDPs), a family of MDPs parameterized by an unobserved context. In CMDPs, an optimal
agent can adapt online by maintaining the posterior belief over contexts. MATE replaces this intractable posterior with a sum-aggregated
memory, leveraging the posterior's permutation invariance to retain provably sufficient expressiveness. Compared to prior
memory architectures, MATE avoids the growing per-step rollout cost of
Transformers and the gradient issues commonly associated with Recurrent Neural Networks (RNNs). Extensive evaluations across diverse
benchmarks demonstrate that MATE provides clear computational advantages while achieving performance comparable to standard sequence-model baselines.
Read the full abstract ▾
All authors Himchan Hwang, Hyeokju Jeong, Gene Chung, Seungyeon Kim, Sangwoong Yoon, Frank Chongwoo Park
Machine Learning
Radial-Angular Geometry for Reliable Update Diagnosis in Noisy-Label Learning
Ningkang Peng, Jingyang Mao, Xiaoqian Peng et al. (+2 more)
Noisy-label methods often estimate sample reliability from forward-space signals such as loss, confidence, or
entropy. These signals indicate whether a sample is difficult to predict, but they do not directly test whether its observed label induces a reliable parameter update. This gap matters because hard clean samples and mislabeled samples can have similar loss while inducing different updates. We recast reliability estimation as diagnosis of the observed-label update. The sample-wise empirical Fisher trace gives a backward-space measure of update energy: for the classifier layer, it factorizes into a prediction-residual term and a feature-sensitivity term, so it captures information beyond scalar loss. Trace, however, is still a radial magnitude signal and cannot decide whether a large update is useful or harmful. We therefore propose Relative
Geometric Conflict (RGC), which compares the observed-label gradient with a reference gradient induced by an EMA teacher. The conflict term helps distinguish large but aligned hard-clean updates from large conflicting updates caused by corrupted labels. Across synthetic and real-world noisy-label
benchmarks, RGC improves hard-clean preservation and accuracy under our evaluation protocol.
Read the full abstract ▾
All authors Ningkang Peng, Jingyang Mao, Xiaoqian Peng, Weiguang Qu, Yanhui Gu
Machine Learning
Progressive Generalization Augmentation with Deeply Coupled RND-PPO and Domain-Prioritized Noise Injection for Robust Crop Management Reinforcement Learning
Wu Yang
Our preliminary experiments on gym-DSSAT maize irrigation tasks revealed that +/-2 degrees C temperature noise causes an 11.9% reduction in economic returns for PPO policies trained under clean conditions - a systematic robustness deficit that existing research has not adequately addressed. This paper tackles three interconnected limitations impeding practical deployment of agricultural RL systems: the trade-off between early-stage learning efficiency and late-stage generalization capability; the naive additive combination of intrinsic and extrinsic rewards in exploration-augmented PPO; and uniform measurement noise injection strategies that disregard empirically validated differential sensitivity across agricultural state variables. We introduce three systematic innovations: Progressive Generalization Augmentation (PGA) implementing a three-phase curriculum (clean training 0-800 episodes, progressive 800-1200, full augmentation 1200-2000); a deeply coupled RND-PPO architecture with dual-channel GAE normalization, progress-decayed intrinsic coefficients, and semantic discretization; and domain-prioritized noise injection with hierarchical activation. Our experimental evaluation demonstrates: 8.43% yield improvement and 16.42% nitrogen use efficiency improvement over SOTA BERT-DQN in Florida; 5.61% yield improvement in Zaragoza (though 3.67% lower economic score due to challenging Mediterranean
climate); and 94.4% vs 80.0% performance retention under combined perturbations. All experiments used 5 random seeds on NVIDIA A100 GPUs with 4.2+/-0.3 hours per run (2000 episodes, 2048-step buffer, 64 mini-batch size).
Read the full abstract ▾
Machine Learning · Cross-listed from cs.MA
Human-Flow Digital Twin for Predicting the Effects of Mobility Introduction on Visitor Circulation
Chiharu Shima, Haruki Yonekura, Fukuharu Tanaka et al. (+2 more)
We propose a framework for predicting the effects of mobility introduction measures using a human-flow digital twin. This digital twin incorporates a multi-
agent simulator that can represent how visitors choose destinations depending on factors such as their current location and the attractiveness of spots. We extract data on how visitors selected destinations with respect to measured pre-intervention human-flow data, inter-spot distances, spot attractiveness, and travel volumes, and use these data to train each
agent's decision model of this simulator. The trained decision model is a function that takes a visitor's current state and surrounding environmental information as input and outputs which spot the visitor will move toward next. By expressing mobility introduction measures as changes to inter-point distances or to spot attractiveness, the framework can reproduce human flows with mobility introduction in the multi-
agent simulator and thereby quantify effects such as changes in visitor counts and circulation. We evaluated the proposed method using human-flow data measured with and without introducing mobility within Wakayama Castle Park in Japan. When reproducing flows with mobility introduction using a multi-layer perceptron decision model, the cosine similarity of the spatial population distribution exceeded 0.7, confirming that the approach can replicate the flow changes caused by the mobility introduction.
Read the full abstract ▾
All authors Chiharu Shima, Haruki Yonekura, Fukuharu Tanaka, Tatsuya Amano, Hirozumi Yamaguchi
Machine Learning
Learning Displacement-Robust Representations for Landslide Early Warning under Rainfall Forecast Uncertainty
Ren Ozeki, Hamada Rizk and Hirozumi Yamaguchi
Rainfall-induced landslides pose a growing risk worldwide as
climate change intensifies extreme rainfall events. To provide sufficient evacuation time, landslide early warning systems (LEWS) for real-time disaster monitoring must estimate near-future landslide risk by integrating observed rainfall with short-term rainfall forecasts from spatio-temporal environmental data streams. Although recent landslide prediction methods have improved predictive performance using statistical and deep learning approaches, most assume accurate rainfall inputs. In operational settings, however, landslide prediction relies on rainfall forecasts, which often contain spatial displacement of rainfall fields due to
forecasting uncertainties. Such displacement can alter local accumulated rainfall and degrade prediction accuracy. To address this challenge, we propose a novel LEWS robust to rainfall field displacement. The key idea is to learn latent representations from rainfall and terrain data that remain stable under displacement in rainfall field motion, enabling reliable geospatial data integration for landslide risk estimation. The landslide prediction model is trained using Rainfall-Motion-Aware Contrastive Learning (RMCL), which introduces temporally correlated rainfall field perturbations to emulate forecast-induced displacement in rainfall-driven spatio-temporal environmental data streams. Experiments were conducted using two years of rainfall and terrain data across Japan, covering 19 regions with landslide events. The proposed system achieved up to 37% higher precision than state-of-the-art baselines. These results demonstrate that modeling rainfall as a moving spatial field and addressing rainfall field displacement during learning significantly improve the reliability of short-term landslide prediction in operational early warning systems.
Read the full abstract ▾
Machine Learning
IVF-TQ: Calibration-Free Streaming Vector Search via a Codebook-Free Residual Layer
Tarun Sharma
Approximate nearest neighbor (ANN) indexes deployed against streaming corpora silently lose recall over weeks. The standard diagnosis is distribution shift, but under shuffled-i.i.d. ingestion – no shift at all – product quantization still degrades -3.8pp at sub-matched bit budgets. The dominant production compression methods (PQ, OPQ, ScaNN) all fit a codebook to an initial sample and reuse it as the database grows by orders of magnitude. This paper presents IVF-TQ, an inverted-file index whose residual compression layer is data-independent: a fixed random rotation followed by a precomputed Lloyd-Max scalar quantizer parameterised only by the bit width b and dimension d. Only the IVF coarse k-means partition is trained. A uniform-over-sphere inner-product error bound depending only on (b, d, delta) provides a structural guarantee no learned-codebook method admits. The same codebook-free design enables an IVF-amplification effect that closes the gap to Extended RaBitQ to within statistical noise (+17.7pp over flat TQ at matched bit budget), and an Adaptive variant that refreshes the partition without touching the compression layer. Across nine controlled
cells (three 10M
datasets, three PQ
memory regimes, three seeds), per-batch PQ codebook retraining never recovers the streaming gap; IVF-PQ streaming stability requires per-
dataset bit-budget tuning, while IVF-TQ holds at one fixed (b, d) configuration on all three
datasets with Delta in [-0.80, +0.56]pp. The contribution is operational: no codebook to train, no per-
dataset bit-budget tuning, no retraining cycle that ever closes the gap.
Read the full abstract ▾
Machine Learning
Self-Supervised Learning for Sparse Matrix Reordering
Ziwei Li, Tao Yuan, Fangfang Liu et al. (+3 more)
Rearranging the rows or columns of a sparse
matrix using an appropriate ordering can significantly reduce fill-ins, i.e., new nonzeros introduced during
matrix factorization, decreasing
memory usage and runtime. However, finding an ordering that minimizes fill-ins is NP-complete. Existing approaches, including graph-theoretic and deep learning methods, rely on surrogate objectives without theoretical guarantees. The Fill-Path
Theorem reveals a direct and intrinsic relationship between fill-in generation and the sparse structure of the
matrix as path triplet inequalities. Here we first employ a multigrid graph network to capture structural information for each vertex. We then derive a triplet sampling strategy based on inequalities. Finally, we introduce an end-max chain loss function to reduce the number of triplets whose predicted scores satisfy these inequalities. Experimental evaluations on the publicly available SuiteSparse
matrix collection demonstrate the superiority of the proposed method in terms of both fill-in reduction and speedup in LU factorization time.
Read the full abstract ▾
All authors Ziwei Li, Tao Yuan, Fangfang Liu, Shuzi Niu, Huiyuan Li, Wenjia Wu
Machine Learning · Cross-listed from cs.CL
MiniGPT: Rebuilding GPT from First Principles
Jibin Joseph
This paper presents MiniGPT, a compact from-scratch implementation of GPT-style autoregressive
language modeling in PyTorch. The aim is to rebuild the core GPT pipeline from first principles after studying the design of nanoGPT by Andrej Karpathy, while keeping the model and training code independently written in a single notebook. MiniGPT implements
token and positional
embeddings, causal multi-head self-
attention, pre-LayerNorm
Transformer blocks, residual connections, feed-forward MLP layers, next-
token cross-
entropy training (teacher forcing), validation tracking, checkpoint selection, and autoregressive text generation. This paper evaluates the implementation on Tiny Shakespeare
dataset using character-level tokenization. A baseline 0.83M-parameter model reaches a validation loss of 1.7236 after 3000 training iterations. A stronger 10.77M-parameter configuration, using a larger context length and improved training settings, reaches a best validation loss of 1.4780 and generates text with recognizable Shakespeare-style dialogue structure. MiniGPT does not introduce a new
language-model architecture. Instead, it documents a clear and reproducible implementation path from raw text to trained character-level generation, including design choices, training behavior, generation quality, and practical limitations.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.AI
Heterogeneous Information-Bottleneck Coordination Graphs for Multi-Agent Reinforcement Learning
Wei Duan, Junyu Xuan, En Yu et al. (+2 more)
Coordination graphs are a central abstraction in cooperative multi-
agent reinforcement learning (MARL), yet existing sparse-graph learners lack a theoretically grounded mechanism to decide which edges should exist and how much information each edge should carry. Current methods rely on heuristic criteria that offer no formal guarantee on the learned
topology, and no principled way to allocate different communication capacities to structurally different
agent relationships. To address this, we propose Heterogeneous Information-Bottleneck Coordination Graphs (HIBCG), which learns a group-aware sparse graph in which both edge existence and message capacity are theoretically justified. With the graph information bottleneck (GIB) serving as the underlying tool, HIBCG first constructs a group-aligned block-diagonal prior that provides a closed-form criterion for edge retention – determining which edges should exist and at what density per group block – and then controls per-
agent feature bandwidth on the resulting
topology, compressing messages to retain only task-relevant content. We prove that the group-aligned prior strictly tightens the variational bound on
topology learning, that the objective decomposes per group block, enabling differential edge control, and that capacity allocation follows a water-filling principle.
Read the full abstract ▾
All authors Wei Duan, Junyu Xuan, En Yu, Xiaoyu Yang, Jie Lu
Machine Learning · Cross-listed from cs.SE
NOETHER: A Constructive Framework for Metamorphic Pattern Discovery from Operator Algebras
Meng Li, Xiaohua Yang, Jie Liu et al. (+1 more)
Context. Metamorphic Testing is recognised in IEEE/ISO software-testing standards and increasingly recommended for AI systems, but its progress is bottlenecked by metamorphic relation (MR) identification: existing approaches (structured frameworks, mining and evolutionary pipelines, LLM-assisted methods, MetaPattern catalogues) share an inductive grounding that leaves three foundational questions open: origin, closure, and transferability. Objective. We propose a framework whose downstream step from program-induced operator
algebra to MetaPattern set is mechanical and provable, while the upstream curation of the
algebra is a stated empirical hypothesis with explicit scope precondition. Method. NOETHER is a two-layer framework. The upstream layer is an eight-block decomposition over recurrent mathematical structures (
symmetry, order, self-adjoint, time-reversal, limit, qualitative-dynamics, method-comparison, relational equivalence). The downstream CONSTRUCT-MP
algorithm produces a MetaPattern set with algebraic-closure (
Theorem 1) and
polynomial-time decidability (
Theorem 2) guarantees. We test the framework on three operator-algebraic domains. Results. On Boltzmann reactor physics NOETHER systematises a prior inductive catalogue; on equivariant ML it derives executable MRs for rotation invariance, adjoint duality, and training-trajectory reversibility; on relational query optimisers it exercises the relational-equivalence block. The central falsifiable prediction (L*-blindness on homogeneity-preserving mutators) holds on the in-scope substrate. The absolute-completeness
conjecture (
Theorem 1') is falsified on PWR core
diffusion via two pairwise-independent counterexamples that identify five Translate-extension dimensions. Conclusion. Induction is relocated from per-program MR sampling to a per-domain algebraic layer; the downstream step is deductive and mechanical.
Read the full abstract ▾
All authors Meng Li, Xiaohua Yang, Jie Liu, Shiyu Yan
Machine Learning · Cross-listed from cs.AI
ADR: An Agentic Detection System for Enterprise Agentic AI Security
Chenning Li, Pan Hu, Justin Xu et al. (+9 more)
We present the Agentic AI Detection and Response (ADR) system, the first large-scale, production-proven enterprise framework for securing AI
agents operating through the Model Context Protocol (MCP). We identify three persistent challenges in this domain: (1) limited observability – existing Endpoint Detection and Response (EDR) tools see file writes but not the
agent reasoning,
prompts, or causal chains linking intent to execution; (2) insufficient robustness – static defenses constrained by pre-defined rules fail to generalize across diverse attack techniques and enterprise contexts; and (3) high detection costs – LLM-based inference is prohibitively expensive at scale. ADR addresses these challenges via three components: the ADR Sensor for high-fidelity agentic telemetry, the ADR Explorer for systematic pre-deployment red teaming and hard-example generation, and the ADR Detector for scalable, two-tier online detection combining fast triage with context-aware
reasoning. Deployed at Uber for over ten months, ADR has sustained reliable detection in production with growing adoption reaching over 7,200 unique hosts and processing over 10,000
agent sessions daily, uncovering hundreds of credential exposures across 26 categories and enabling a shift-left prevention layer (97.2% precision, 206 detected credentials). To validate the approach and enable community adoption, we introduce ADR-Bench (302 tasks, 17 techniques, 133 MCP servers), where ADR achieves zero false positives while detecting 67% of attacks – outperforming three state-of-the-art baselines (ALRPHFS, GuardAgent, LlamaFirewall) by 2–4x in F1-score. On AgentDojo (public
prompt injection
benchmark), ADR detects all attacks with only three false alarms out of 93 tasks.
Read the full abstract ▾
All authors Chenning Li, Pan Hu, Justin Xu, Baris Ozbas, Olivia Liu, Caroline Van, Manxue Li, Wei Zhou, Mohammad Alizadeh, Pengyu Zhang, KK Sriramadhesikan, Ming Zhang
Machine Learning
FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics
Qiran Zou, Hou Hei Lam, Wenhao Zhao et al. (+11 more)
AI research
agents accelerate ML research by automating hypothesis generation, experimentation, and empirical refinement. Existing
agent strategies range from greedy hill-climbing to tree search and evolutionary optimization, yet which strategy choices drive performance remains unclear. Answering this question requires a
benchmark that separates
agent strategy (e.g., search
topology) from execution infrastructure (e.g., code editor), so that performance differences are attributable to strategy rather than infrastructure, and that provides process-level metrics beyond final scores to analyze exploration behaviors. Existing
benchmarks offer limited support. We propose FML-Bench, a
benchmark of 18 fundamental ML research tasks across 10 domains that separates
agent strategy from execution infrastructure and defines 12 process-level behavioral metrics. Evaluating six representative
agents, we find that: (1) strategy complexity alone does not guarantee strong performance: a simple greedy hill-climber nearly matches the best-performing tree-search
agent, both well above the remaining
agents; (2) our analysis suggests this pattern relates to improvement opportunity structure: greedy search tends to be more effective when opportunities are dense, while tree-search and evolutionary strategies tend to be more effective when opportunities are sparse; an adaptive
agent built on this insight switches to broader exploration upon detecting improvement stagnation and outperforms the other six
agents, lending initial support to this observation; and (3) process-level analysis reveals that early convergence and directionally focused exploration are significantly associated with final performance, while solution diversity and compute cost are not. Our
benchmark is available at: https://github.com/qrzou/FML-bench.
Read the full abstract ▾
All authors Qiran Zou, Hou Hei Lam, Wenhao Zhao, Tingting Chen, Yiming Tang, Samson Yu, Yingtao Zhu, Srinivas Anumasa, Zufeng Zhang, Tianyi Zhang, Chang Liu, Zhengyao Jiang, Anirudh Goyal, Dianbo Liu
Machine Learning
Learning Fill-in Reduction Ordering via Graph Policy Optimization for Sparse Matrices
Ziwei Li, Shuzi Niu, Huiyuan Li et al. (+2 more)
Matrix reordering in large sparse solvers seeks a permutation that minimizes factorization fill-in to reduce
memory and computation. Because the minimum fill-in ordering problem is NP-complete and fill-in is implicit in the sparsity pattern, graph-theoretic heuristics are used. Existing reinforcement learning methods either ignore sparsity patterns–missing the global fill-in–or lack local exact fill-in feedback. We propose a graph policy optimization method, modeling fill-ins from global and local views: both the policy and value networks use a multi-hop graph neural backbone to embed global fill-in; the policy further interacts with symbolic factorization over graphs to extract local, step-level fill-ins, and the resulting feedback is aligned with the value network via an adaptive saturation function to improve convergence. On the SuiteSparse
Matrix Collection, our method achieves mean reductions of 29.3 in fill-ins and 31.3 in peak
memory usage over state-of-the-art baselines.
Read the full abstract ▾
All authors Ziwei Li, Shuzi Niu, Huiyuan Li, Tao Yuan, Wenjia Wu
Machine Learning
MasFACT: Continual Multi-Agent Topology Learning via Geometry-Aware Posterior Transfer
Xuefei Wang, Jialu Wang, Fengbo Zhang et al. (+6 more)
Multi-
agent systems (MAS) powered by large
language models (LLMs) have emerged as a powerful paradigm for complex problem solving, where performance critically depends on the underlying inter-
agent communication
topology. However, existing
topology generation methods mainly optimize for isolated tasks, while real-world deployments involve streams of evolving tasks, requiring previously effective collaboration patterns to be retained and reused rather than rediscovered or overwritten. We identify a previously underexplored failure mode,
topology forgetting, in which adapting to new tasks shifts the
topology generator away from communication structures required by earlier tasks. This issue stems from cross-task misalignment in both
agent-level functional semantics and relational communication structures. To address this challenge, we propose \textbf{MasFACT}, a geometry-aware posterior transfer framework that preserves and reuses historical collaboration knowledge as transferable
topology priors. We transfer these priors across task-specific
agent spaces through Fused Gromov-Wasserstein optimal transport and perform PAC-Bayes-guided conservative posterior adaptation to balance task-specific plasticity with structural stability. Experiments across class-, domain-, and task-level continual settings demonstrate that MasFACT consistently improves average accuracy while reducing
topology forgetting compared to strong
topology generation and replay-based baselines, and can be seamlessly integrated with different MAS
topology generators.
Read the full abstract ▾
All authors Xuefei Wang, Jialu Wang, Fengbo Zhang, Yihan Hu, Di Zhang, Yutong Ye, Yikun Ban, Jun Han, Ruijie Wang
Machine Learning · Cross-listed from cs.CY
Position: Age Estimation Models Do Not Process Biometric Data
Nikita Marshalkin
When a neural network estimates someone's age from a photograph, does it process biometric data? The answer depends on whether identity-discriminative representations arise within the network during inference, a question that may seem trivial to ML researchers but triggers consent requirements under GDPR, statutory damages under BIPA, or high-risk AI classification under the EU AI Act. Yet no regulatory guidance addresses it. This position paper provides empirical evidence: 14 models evaluated across 3 face verification
benchmarks show age
estimators fall orders of magnitude short of identification thresholds. Age estimation models cannot identify individuals. We call on researchers to provide transparency about what systems store and can do, and on regulators to distinguish transient processing from template storage.
Read the full abstract ▾
Machine Learning
Olivia: Harmonizing Time Series Foundation Models with Power Spectral Density
Jingru Fei, Kun Yi, Alex Xing Wang et al. (+3 more)
Time series foundation models rely on large-scale pretraining over diverse
datasets across domains, yet their heterogeneity in temporal patterns could hinder the effectiveness of training and learning transferable time series representations. Inspired a fundamental concept, normalized power spectral density (PSD) in signal processing, we assume harmonizing
datasets via PSDs in the spectral domain could reduce mismatches and enhance pretraining. We then go beyond the direct intractable minimization optimization and innovatively reformulate it as a principled harmonization approach. Specifically, we propose Harmonizer, a module that reshapes spectral structures and implicitly harmonizing PSDs across
datasets, which theoretically corresponds to a shared reparameterization of second-order temporal correlations. Our theoretical analysis further reveals
token interactions with Harmonizer can be efficiently mediated by a compact set of resonators, motivating a HarmonicAttention design that performs self-
attention in a low-dimensional interaction space. Then, we propose Olivia, a novel time series foundation model built upon these harmonization mechanisms. Extensive experiments on two large-scale
benchmarks (TSLib and GIFT-Eval) and extra 6
datasets from GluonTS, demonstrate Olivia consistently achieves state-of-the-art performance under zero-shot, few-shot, and full-shot
forecasting scenarios. Our code is available at https://github.com/TSTS13/Olivia.
Read the full abstract ▾
All authors Jingru Fei, Kun Yi, Alex Xing Wang, Qingsong Wen, Xiangxiang Zhu, Wei Fan
Machine Learning
Bridging the Gap between Sparse Matrix Reordering and Factorization: A Deep Learning Framework for Fill-in Reduction
Ziwei Li, Tao Yuan, Shuzi Niu et al. (+1 more)
Sparse
matrix reordering can significantly reduce the fill-in during
matrix factorization, thereby decreasing the computational and storage requirements in sparse
matrix computations. Finding a minimal fill-in ordering is known to be an NP-hard problem. Moreover, there is a paradox:
matrix reordering is applied before
matrix factorization, but fill-ins that
matrix reordering methods aim at are generated from
matrix factorization. To bridge the gap between reordering and factorization, we propose a deep learning framework to minimize a fill-in surrogate function based on spectral
embedding. First, we employ a multi-grid-like GNN architecture to learn to approximate the smallest eigenvectors of its graph
Laplacian matrix, i.e. spectral
embedding, and capture the global structural information of the
matrix. Then, another multi-grid-like GNN architecture is used to minimize the potential space where fill-in can occur based on the rank distribution. Experimental results indicate that our approach achieves competitive performance compared with traditional graph-theoretic
algorithms and deep learning methods.
Read the full abstract ▾
All authors Ziwei Li, Tao Yuan, Shuzi Niu, Huiyuan Li
Machine Learning · Cross-listed from cond-mat.mtrl-sci
Causal Anomaly Detection for Lithium-Ion Battery Degradation
Dieter W. Heermann and Hagen Heermann
Reliable early detection of lithium-ion battery degradation requires health indicators that are physically interpretable and computable from routine cycler telemetry without access to the degradation region. We introduce CausalHealth, a framework that applies causal graph discovery and $k$-nearest-neighbour transfer
entropy to per-cycle voltage, current, temperature, and resistance time series, and organises twelve resulting anomaly scores into three signal-class bundles (Magnitude-shift, Predictive-residual, Complexity-
entropy) – with Isolation Forest reported separately as it falls below the bundle reliability threshold – to characterise detection sensitivity across ten commissioning fractions (5–30\,%). The Magnitude-shift class achieves 100\,% detection across all seven tested
cells spanning LFP (MIT–Stanford MATR) and LCO (NASA PCoE, CALCE CS2) chemistries, with a lead time of up to 402 cycles before conventional capacity-threshold failure on gradual-fade
cells. A Reliability-Weighted Master Health Index (RWMHI) – a cross-bundle fusion of five high-reliability detectors weighted by inverse coefficient of variation – improves lead time by 15–52 cycles over the class median on long-lived
cells while maintaining 100\,% detection. Validation against electrochemical impedance spectroscopy on an NMC prismatic
cell provides independent physical grounding: transfer
entropy $\mathrm{TE}(R \!\to\! V)$ correlates with charge-transfer resistance $R_{\mathrm{ct}}$ (pooled $r = +0.990$; temperature-controlled partial $r = +0.898$), and an Arrhenius analysis of both quantities yields an activation energy consistent with published NMC charge-transfer kinetics. These results are evaluated on seven
cells across three
benchmark datasets.
Read the full abstract ▾
Machine Learning
Leveraging Error Diversity in Group Rollouts for Reinforcement Learning
Wenpu Liu, Yuqi Xu, Weichu Xie et al. (+8 more)
Reinforcement Learning from Verifiable Rewards (RLVR) typically samples multiple responses per
prompt and assigns binary rewards based on individual correctness, yet the collective structure of the group output, specifically the distribution of errors, is largely discarded. We identify this as a missed opportunity: empirical analysis reveals that error diversity within a group is a strong predictor of training success, with problems eliciting diverse wrong answers benefiting substantially more from RLVR than those producing homogeneous failures. Motivated by this observation, we propose Error Diversity Advantage Shaping (EDAS), a lightweight,
algorithm-agnostic technique that modulates the advantage signal for incorrect rollouts based on intra-group error diversity. EDAS amplifies penalties for dominant, repeated errors and attenuates penalties for rare, exploratory ones, thereby encouraging the model to maintain diverse
reasoning paths and discouraging error perseveration. Crucially, EDAS operates as a simple post-hoc adjustment that can be seamlessly integrated into any RLVR
algorithm. We validate EDAS on top of several mainstream RLVR methods across a series of models and seven challenging math
benchmarks, demonstrating consistent improvements. Notably, EDAS yields an average improvement of 6.29 points over DAPO on Qwen3-8B across seven
benchmarks, confirming that exploiting the latent information in group rollouts is a broadly effective strategy for strengthening RLVR.
Read the full abstract ▾
All authors Wenpu Liu, Yuqi Xu, Weichu Xie, Yongfu Zhu, Shuai Dong, Ziyue Wang, Wenqi Shao, Xiaoying Zhang, Tong Yang, Nan Duan, Jiaqi Wang
Machine Learning · Cross-listed from hep-lat
Noise scheduling and linear dynamics in diffusion models on Lie groups
Javad Komijani
We investigate the role of the noise schedule in
diffusion processes on Lie groups, with particular emphasis on applications to
lattice gauge theory. We show that a specific noise schedule leads to a linear decay of the expectation value of the Wilson action as a function of
diffusion time. We compare this with Euclidean
diffusion models, where such behavior requires an explicitly designed drift term, while in the Lie-group setting it arises naturally.
Read the full abstract ▾
Machine Learning
Learning Higher-Order Interactions from Incomplete Spatiotemporal Data with Mixtures of Hypergraph Laplacians
Keshu Wu, Sixu Li, Zihao Li et al. (+3 more)
Traffic sensor data exhibit strong temporal and cross-sensor dependencies, yet long gaps erase both a sensor's local history and the joint observations needed to learn its relationships with other sensors. We introduce MoHL, a mixture of hypergraph
Laplacians that learns group-wise relational structure directly from incomplete spatiotemporal data. MoHL discovers overlapping sensor groups from observed co-variation, constructs a
Laplacian component for each interaction order, and learns their mixture through holdouts that mimic the missingness pattern. By sharing weights across group-induced couplings, MoHL pools scattered evidence instead of estimating each coupling independently. We characterize the induced weight-sharing structure and analyze the conditions under which sharing reduces estimation risk. Experiments on three traffic networks show that higher-order groups are most beneficial under long gaps, while support-matched comparisons demonstrate the advantage of shared weights when relational evidence is scarce. MoHL achieves the lowest error among the evaluated baseline methods under day-long gaps, with orders of magnitude fewer learned parameters than deep imputers. These results show how group-wise sharing can sustain reliable reconstruction precisely when missingness depletes the evidence needed to learn cross-sensor dependence.
Read the full abstract ▾
All authors Keshu Wu, Sixu Li, Zihao Li, Zhiwen Fan, Xiaopeng Li, Yang Zhou
Machine Learning · Cross-listed from cs.CL
Weak-to-Strong Elicitation via Mismatched Wrong Drafts
Wei Deng
We consider whether off-policy experience from a smaller, weaker model can elicit capability in a stronger learner that on-policy RL fine-tuning (e.g., GRPO) does not reach. We find that injecting mathematically wrong drafts from a smaller but more domain-trained model – mismatched to the current problem – into a stronger learner's GRPO context consistently outperforms standard on-policy GRPO on held-out MATH-500 and out-of-distribution AIME 2025/2026. Concretely, we use Mathstral-7B as the learner, Qwen2.5-Math-1.5B as the draft model, 8.8K Level 3–5 MATH problems (with MATH-500 held out), and train with Dr. GRPO. Mismatch is an active ingredient: shuffling drafts to mismatched problems while holding everything else constant yields $+1.62$pp on MATH-500 (greedy pass@1) over the matched-wrong variant ($n=10$ seeds, $p=0.0015$, Welch's $t$). In fact, the mismatched-wrong variant leads all other variants we tested on MATH-500 across both greedy pass@1 and sampling pass@$k$. On out-of-distribution AIME 2025 and 2026, the mismatched-wrong variant uniquely lifts pass@$k$ above both Mathstral-7B (in its native [INST] format) and the Qwen2.5-Math-1.5B draft model at every sample budget from $k=1$ to $k=1024$ across 2 seeds ($+14.2$pp on 2025 and $+9.0$pp on 2026 at pass@1024 over Mathstral-7B), and at pass@1024 also leads no-draft, matched-wrong, and mismatched-correct variants on both years. All variants use the same
prompt with no draft injection at test time. The recipe – trained on a single GPU with no SFT, no reward models, no synthesized data, and no produce-critique-revise inner loop – reaches 71.98% MATH-500 on Mathstral-7B-v0.1, the highest published result on this model to our knowledge, surpassing the heavier WizardMath pipeline at 70.9% on full MATH (SFT + PPO with process/instruction reward models).
Read the full abstract ▾
Machine Learning · Cross-listed from q-fin.PM
Deep Reinforcement Learning Framework for Diversified Portfolio Management Across Global Equity Markets
Kamil Kashif and Robert Ślepaczuk
This study develops and evaluates a deep reinforcement learning framework for dynamic portfolio allocation across global equity
markets. The Soft Actor-Critic
algorithm is used to learn continuous portfolio weights within a Markov Decision Process, incorporating transaction costs, turnover penalties, and diversification constraints into the reward function. Five model configurations are compared, varying in reward formulation, policy structure (flat versus hierarchical Dirichlet), portfolio constraints, and temporal encoder (LSTM versus
Transformer), and evaluated via walk-forward optimization across sixteen out-of-sample folds spanning 2003-2026 on the Nasdaq-100, Nikkei 225, and Euro Stoxx 50. Results show that RL strategies achieve competitive risk-adjusted performance primarily in the Euro Stoxx 50, where statistically significant abnormal returns are observed, but the central hypothesis is only partially confirmed: no strategy achieves statistically significant excess returns relative to Buy and Hold under HAC-robust inference across all
markets. Regime analysis reveals that RL adds the most value during periods of elevated uncertainty, while ensemble aggregation across
markets improves risk-adjusted performance and confirms the benefits of geographic diversification.
Read the full abstract ▾
Machine Learning
Compress the Context, Keep the Commitments: A Formal Framework for Verifiable LLM Context Compression
Natalia Trukhina and Vadim Vashkelis
LLM context is not just
tokens; it is a set of commitments. Long-running conversations accumulate goals, constraints, decisions, preferences, tool results, retrieved evidence, artifacts, and safety boundaries that future responses must preserve. Existing context-management methods reduce length through truncation,
retrieval, summarization,
memory systems, or
token-level
prompt compression, but they rarely specify which semantic commitments must survive compression or how their preservation should be measured. We propose Context Codec, a commitment-level framework for compressing
prompts and chat histories. Context Codec represents dialogue state as typed, source-grounded semantic atoms with canonical identity, equivalence, conflict, confidence, risk, and evidence spans. It separates five concerns - extraction, normalization, representation, rendering, and verification - and introduces metrics for Critical Atom Recall, Weighted Atom Recall, Commitment Density, and round-trip recoverability. It also defines a taxonomy of semantic compression errors, a concrete normalization procedure, conservative fallback rules for low-confidence and safety-critical atoms, and Context Compression
Language (CCL), an ASCII-first compact rendering of canonical JSON atoms. In a small diagnostic study, CCL-Core occupies a useful middle ground between structured prose and JSON: more explicit and auditable than prose, usually more compact than JSON, and less risky than heavily minified notation. The result is not a claim that shorthand solves compression, but a framework for making context compression verifiable: compress the conversation, keep the commitments.
Read the full abstract ▾
Machine Learning
DISA: Offline Importance Sampling for Distribution-Matching LLM-RL
Shaobo Wang, Yujie Chen, Yafeng Sun et al. (+9 more)
Modern
reasoning agents are increasingly evaluated on their ability to generate multiple valid solution paths, plans, or tool-use traces for a given input. Standard reward-maximizing RL tends to collapse onto the most easily reinforced high-reward mode, whereas distribution-matching RL aims to allocate probability mass across the entire reward-shaped solution set. Achieving this objective requires computing a
prompt-dependent partition function over the trajectory space. Because existing distribution-matching methods learn this partition function online alongside the policy, calibration errors in the partition function directly distort policy updates and remain impossible to diagnose independently. We introduce DISA, short for Decoupled Importance-Sampled Anchoring, which moves this calibration problem outside the RL loop. DISA draws proposal trajectories offline, estimates the partition function via importance sampling, and freezes the resulting partition-function estimate before policy optimization begins. This decoupling preserves the distribution-matching objective while strictly separating partition-function estimation from policy learning in data, gradients, loss, and diagnostics. Empirically, on two open-weight backbones across six math and three code
benchmarks, DISA matches or exceeds the online-coupled distribution-matching baseline FlowRL, outperforms rewardmaximization baselines GRPO and GSPO on math averages, and exceeds LoRASFT
distillation by up to 13.8 Mean@8 points on the same offline trajectories. An LLM-as-judge evaluation further shows that DISA retains substantially more strategy-level diversity than reward-maximization baselines, and sensitivity studies on the proposal strength and inverse temperature follow the bias-variance pattern predicted by the analysis.
Read the full abstract ▾
All authors Shaobo Wang, Yujie Chen, Yafeng Sun, Wenjie Qiu, Zhihui Xie, Sihang Li, Yucheng Li, Huiqiang Jiang, Xingzhang Ren, Xuming Hu, Dayiheng Liu, Linfeng Zhang
Machine Learning
Step-wise Rubric Rewards for LLM Reasoning
Weichu Xie, Haozhe Zhao, Wenpu Liu et al. (+15 more)
Reinforcement Learning with Verifiable Rewards (RLVR) improves
language-model
reasoning using final-answer correctness, but lacks supervision for intermediate steps. Rubric-based methods such as Rubrics as Rewards (RaR) provide finer-grained criteria, yet collapse rubric scores into a response-level scalar. This loses criterion-level structure, rewards flawed steps in correct responses, penalizes valid steps in incorrect responses, and can encourage unbounded self-correction. In 1,000 sampled problems, 18.2% of steps in answer-correct responses are wrong, while 49.9% of steps in answer-incorrect responses are correct. We propose Step-wise Rubrics as Rewards (SRaR), which (i) uses an LLM judge to assign rubric items to
reasoning steps, (ii) normalizes step scores across rollouts to focus learning on quality differences, and (iii) combines step-wise and outcome rewards through a decoupled advantage
estimator that preserves a stable outcome baseline. We also construct a 16K-problem rubric
dataset by contrastively distilling criteria from verified correct and flawed
reasoning paths. Across six math
benchmarks, SRaR improves average accuracy over RaR by 3.57 and 2.75 points on Qwen3-8B-Non-Thinking and Qwen3-32B-Non-Thinking, respectively. On AIME 2025, it increases the Faithful
Reasoning Rate from 34.5% to 46.7% and reduces self-correction looping from 48.1% to 26.5%. SRaR also achieves the best results on three of four code-generation
benchmarks.
Read the full abstract ▾
All authors Weichu Xie, Haozhe Zhao, Wenpu Liu, Yongfu Zhu, Liang Chen, Minghao Ye, Zirong Chen, Yuqi Xu, Shuai Dong, Ziyue Wang, Xinbo Xu, Kean Shi, Ruoyu Wu, Xiaoying Zhang, Wenqi Shao, Baobao Chang, Nan Duan, Jiaqi Wang
Machine Learning
LEAP: Learnable End-to-End Adaptive Pruning of Large Language Models
Mohammad Mozaffari, Younes Hourri, Mohammad Rastegari et al. (+1 more)
Unstructured sparsity is now natively accelerated by recent GPU kernels and dataflow hardware, shifting the bottleneck from inference execution to the
pruning algorithm. State-of-the-art methods for unstructured LLM
pruning are layer-wise surrogates derived from the Optimal
Brain Surgeon principle, and they sacrifice end-to-end accuracy, especially under aggressive sparsity. End-to-end alternatives such as MaskLLM and PATCH show that learnable masks can close this gap, but their categorical-over-patterns parameterization scales with the number of valid masks per row and does not port to the unstructured setting. We introduce LEAP, which replaces this intractable parameterization with a per-weight Bernoulli-via-Gumbel-sigmoid relaxation that makes end-to-end unstructured mask learning tractable. Across five LLM families from 0.5B to 8B parameters at 50% and 60% sparsity, LEAP improves six-task average zero-shot accuracy by +2.59 points on average over ADMM, the best layer-wise baseline in our sweep.
Read the full abstract ▾
All authors Mohammad Mozaffari, Younes Hourri, Mohammad Rastegari, Mahyar Najibi
Machine Learning
UNR-Explainer: Counterfactual Explanations for Unsupervised Node Representation Learning Models
Hyunju Kang, Geonhee Han and Hogun Park
Node representation learning, such as Graph Neural Networks (GNNs), has emerged as a pivotal method in machine learning. The demand for reliable explanation generation surges, yet unsupervised models remain underexplored. To bridge this gap, we introduce a method for generating counterfactual (CF) explanations in unsupervised node representation learning. We identify the most important subgraphs that cause a significant change in the k-nearest neighbors of a node of interest in the learned
embedding space upon perturbation. The k-nearest neighbor-based CF explanation method provides simple, yet pivotal, information for understanding unsupervised downstream tasks, such as top-k link prediction and clustering. Consequently, we introduce UNR-Explainer for generating expressive CF explanations for Unsupervised Node Representation learning methods based on a Monte Carlo Tree Search (MCTS). The proposed method demonstrates superior performance on diverse
datasets for unsupervised GraphSAGE and DGI.
Read the full abstract ▾