Machine Learning
Can Transformers Learn to Verify During Backtracking Search?
Yin Jun Phua, Tony Ribeiro, Tuan Nguyen et al. (+1 more)
Backtracking search underlies classical constraint solvers, planners, and
theorem provers. Recent
transformer-based
reasoning systems explore search trees over their own intermediate steps. A common training recipe fits an autoregressive next-
token loss on offline solver traces. The model's input at each step is a cumulative trace of all prior decisions. The optimal continue-or-backtrack predictor depends only on the current search state, since two trajectories reaching the same state admit the same viable continuations. We show that decoder-only
transformers trained on cumulative traces fail this requirement in two ways: the trace can scatter state features across many positions (scattered
retrieval), and the predictor can condition on the trajectory rather than the state (history
entanglement). We address scattered
retrieval with localization, a trace-level fix that rewrites each decision block to expose state features locally. We address history
entanglement with Selective State
Attention (SSA), a fixed
attention mask that enforces state-based decisions structurally without modifying training data, objective, or parameters. We focus on reactive verification, after propagation has exposed a contradiction. We test SSA on 3-SAT, graph coloring, Blocks World, and backtracking parsing. On same-state pairs that differ only in prior history, SSA emits identical decisions while a cumulative-trained causal baseline does not. Our contribution is a diagnostic of
transformer behavior on serialized trajectory data, paired with a structural fix. Pretrained
language models that search over their own
reasoning steps may face the same failure. Our analysis opens up inference-time context clearing as a candidate way to apply the same isolation without retraining.
Read the full abstract ▾
All authors Yin Jun Phua, Tony Ribeiro, Tuan Nguyen, Katsumi Inoue
Machine Learning
Survive or Collapse: The Asymmetric Roles of Data Gating and Reward Grounding in Self-Play RL
Sophia Xiao Pu, Zhaotian Weng, Chengzhi Liu et al. (+4 more)
Self-play reinforcement learning trains
language models on their own generated tasks, co-evolving a proposer and solver without human labels. Recent systems report strong
reasoning gains, but collapse and instability are widely observed and poorly understood. The dominant response treats this as a reward-design problem. We argue instead that self-play stability is governed by two distinct levers: a data-level gate that decides which proposer-generated tasks enter the training pool, and the reward signal that updates the policy on tasks already admitted. Through controlled experiments on a Python output-prediction task and a deterministic-DSL twin task that strips pretraining priors, output ambiguity, and executor noise, we find the two levers are asymmetric. A strict gate is sufficient for stability under every reward variant we test, including a self-consistency reward with no access to ground truth; while no reward variant is sufficient once the gate is removed. This asymmetry exposes a counter-intuitive coupling we call the Grounded Proposer Paradox: a proposer with ground-truth access accelerates collapse faster than an ungrounded one when paired with a self-consistency solver, by concentrating training on clean tasks that form the fastest path to a spurious self-consistent attractor. Replacing the binary gate with a continuous strictness parameter $\varepsilon$ further reveals a two-stage phase transition: training-side metrics decouple at low $\varepsilon$, while validation accuracy holds until $\varepsilon$ is much higher. Data-level gating, not reward calibration, is the binding constraint on self-play stability.
Read the full abstract ▾
All authors Sophia Xiao Pu, Zhaotian Weng, Chengzhi Liu, Jayanth Srinivasa, Gaowen Liu, William Yang Wang, Xin Eric Wang
Machine Learning · Cross-listed from eess.SY
Kernel-Based Safe Exploration in Deep Reinforcement Learning
Rupak Majumdar, Nikhil Singh and Sadegh Soudjani
Safety has been a major concern when deploying deep reinforcement learning
algorithms in the real world. A promising direction that ensures that the learned policy does not visit unsafe regions is to learn a barrier function along with the policy. A barrier is a function from states to reals that assigns low values to the initial states, high values to the unsafe states, and decreases in expectation on each transition; such a function can be used to bound the probability of reaching unsafe states. Previous attempts learned a barrier function directly from exploration data, but this required either large amounts of data or restrictions on the system dynamics. In this paper, we show how kernel
embeddings can be used to learn barrier functions during deep reinforcement learning for
stochastic systems with unknown dynamics. Our
algorithm, kernel-based safe exploration (KBSE), learns an optimal policy and a barrier simultaneously during exploration. The barriers are computed iteratively, represented as conditional mean
embeddings, and provide better probabilistic safety guarantees with more exploration. The exploration
algorithm uses the learned barrier functions to identify safety violations. In the case of violation, it intervenes to modify the unsafe action to a safe action, thereby ensuring that the exploration is restricted to actions that bound the probability of reaching unsafe states. We evaluate KBSE on several complex continuous control
benchmarks. Experimental results establish our new
algorithm to be suitable for synthesizing control policies that are probabilistically safe without degradation in reward accumulation.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.AI
Skill Weaving: Efficient LLM Improvement via Modular Skillpacks
Zhuo Li, Guodong Du, Zesheng Shi et al. (+5 more)
Large
language models increasingly require specialization across diverse domains, yet existing approaches struggle to balance multi-domain capacities with strict
memory and inference constraints. In this work, we introduce SkillWeave, a modular improvement framework that enables LLMs to specialize under fixed
memory budgets. SkillWeave partitions full capabilities of a general-purpose model into skillpacks – lightweight, domain-specific delta modules – that reorganize and refine the model's internal knowledge. For efficient deployment, SkillWeave integrates SkillZip to compress skillpacks into compact and inference-ready format, enabling strong multi-domain performance with low-latency execution. On multi-task and agentic
benchmarks, a 9B SkillWeave model outperforms several baselines and even surpasses a 32B monolithic LLM, while achieving up to 4x speedup.
Read the full abstract ▾
All authors Zhuo Li, Guodong Du, Zesheng Shi, Weiyang Guo, Weijun Yao, Yuan Zhou, Jiabo Zhang, Jing Li
Machine Learning · Cross-listed from cs.CV
OSS: Open Suturing Skills Vision-Based Assessment Challenge 2024-2025
Hanna Hoffmann, Setareh Bady, Claas de Boer et al. (+54 more)
Achieving high levels of
surgical skill through effective training is essential for optimal patient outcomes. Automated, data-driven skill assessment holds significant potential to improve
surgical training. While machine learning-based methods are increasingly popular for assessing skills in minimally invasive surgery, their application to open surgery remains limited. We present the results of a dedicated MICCAI challenge designed to
benchmark and advance
vision-based skill assessment in open surgery. The challenge
dataset comprises
videos of an open suturing training task recorded with a static GoPro camera in a dry-lab setting, with instrument trajectories available in addition to the primary
video modality. The OSS Challenge was hosted over two consecutive years, comprising two and three independent tasks, respectively: (1) classifying skill level into four classes, (2) predicting the full Objective Structured Assessment of Technical Skills across eight categories, and (3) tracking hands and
surgical tools. Participants submitted diverse solutions including deep learning-based
video models, tracking-driven methods, and hybrid approaches. General-purpose spatiotemporal
video models consistently achieved the strongest performance, though conceptually diverse approaches reached competitive levels when well-executed. Predicting fine-grained OSATS scores remains challenging but benefits substantially from increased training data. Keypoint tracking proves difficult given frequent occlusions and out-of-frame instances, limiting current applicability for motion-based skill analysis. This work
benchmarks innovative and diverse solutions for
surgical skill assessment, highlighting both the promise and current limitations of
video-based evaluation in open surgery and identifying critical directions for advancing automated skill assessment toward clinical impact.
Read the full abstract ▾
All authors Hanna Hoffmann, Setareh Bady, Claas de Boer, Max Kirchner, Jan Egger, Rainer Röhrig, Frank Hölzle, Lennart Johannes Gruber, Kunpeng Xie, Marlon Neuhaus, Victor Alves, Guilherme Barbosa, Leonardo Barroso, João Carvalho, Hao Chen, Gabriella d'Albenzio, André Ferreira, Nuno Gomes, Yuichiro Hayashi, Kousuke Hirasawa, Rebecca Hisey, Seungjae Hong, Seoi Jeong, Tiago Jesus, Daehong Kang, Satoshi Kasai, Shunsuke Kikuchi, Takayuki Kitasaka, Satoshi Kondo, Hyoun-Joong Kong, Youngbin Kong, Atsushi Kouno, Shlomi Laufer, Kyu Eun Lee, Bining Long, Nooshin Maghsoodi, Hiroki Matsuzaki, Evangelos Mazomenos, Ori Meiraz, Kensaku Mori, Marina Music, Masahiro Oda, Roi Papo, Jieun Park, Rafael Piexoto, Saeid Rezaei, Mariana Ribeiro, Soyeon Shin, Yang Shu, Idan Smoller, Danail Stoyanov, Yihui Wang, Xinkai Zhao, Sebastian Bodenstedt, Isabel Funke, Stefanie Speidel, Behrus Hinrichs-Puladi
Machine Learning
Reinforced Graph of Thoughts: RL-Driven Adaptive Prompting for LLMs
Manuel Noah Riesen and Peter Alfred von Niederhäusern
Graph of Thoughts (GoT), a generalized form of recent prompting paradigms for large
language models (LLMs), has been shown to be useful for elaborate problem solving. By executing a graph of operations, thoughts of the LLM are structured as an arbitrary graph, forming the actual graph of thoughts. Originally, the graph of operations is defined manually, which requires in-depth knowledge about the solution of the problem to solve. Such a static graph of operations is rigid and therefore lacks adaptability. We propose Reinforced Graph of Thoughts (RGoT), an automated approach to the GoT prompting paradigm that leverages reinforcement learning (RL) to adaptively generate a graph of operations from a human-defined set. Results indicate that, under certain constraints, it is possible to construct graphs of operations adaptively to the task's complexity in an automated way.
Read the full abstract ▾
Machine Learning
Bandit Convex Optimization with Gradient Prediction Adaptivity
Shuche Wang, Adarsh Barik and Vincent Y. F. Tan
Bandit convex optimization (BCO) is a fundamental online learning framework with partial feedback, where the learner observes only the loss incurred at the chosen decision point in each round. In this work, we investigate whether optimistic gradient predictions can improve worst-case regret guarantees in a prediction-adaptive manner. Specifically, given gradient predictions $m_t$, we seek regret bounds that scale with the cumulative prediction error $S_T=\sum_{t=1}^T \|\nabla f_t(x_t)-m_t\|^2.$ We first establish a negative result: under the single-point feedback protocol, an unavoidable $\Omega(\sqrt{T})$ regret lower bound persists even when $S_T=o(T)$, showing that the variance of gradient estimation fundamentally obscures the benefit of accurate predictions. To overcome this barrier, we propose Two-Point Variance-Reduced Optimistic Gradient Descent (TP-VR-OPT) for the two-point feedback setting. The key idea is a novel variance-reduced gradient
estimator whose variance scales with the prediction error rather than the gradient norm. This yields a regret bound of $O\big(\sqrt{d\,\mathbb{E}[S_T]}\big),$ where $d$ is the decision dimension. Complementing this result, we establish an information-theoretic lower bound that scales as $\Omega(\sqrt{\mathbb{E}[S_T]})$, providing a fundamental characterization of the best achievable prediction-adaptive regret and showing that TP-VR-OPT is optimal up to a factor of $\sqrt d$. We further develop adaptive variants that eliminate the need for prior knowledge of $\mathbb{E}[S_T]$ or the horizon $T$, and extend our framework to non-stationary environments, establishing dynamic regret guarantees that adapt simultaneously to the cumulative prediction error and the comparator path length.
Read the full abstract ▾
Machine Learning
From Sequential Nodes to GPU Batches: Parallel Branch and Bound for Optimal $k$-Sparse GLMs
Jiachang Liu and Andrea Lodi
GPUs have significantly accelerated first-order methods for large-scale optimization, especially in continuous optimization. However, this success has not transferred cleanly to problems with discrete variables, combinatorial structure, and nonlinear objectives, such as certifying optimal solutions for cardinality-constrained generalized linear models. Major challenges include the sequential processing of heterogeneous nodes in branch and bound (BnB) and frequent data movement between the CPU and GPU. We propose a simple, generic, and modular CPU–GPU framework that processes multiple BnB nodes in batches on GPUs. The framework is built around a small set of GPU-efficient routines and uses padding together with lightweight custom kernels to handle irregular node data structures. Experiments show one to two orders of magnitude speedups and zero optimality gap on challenging instances. The framework can also be extended to collect the entire Rashomon set, enabling downstream statistical analysis such as variable-importance analysis and model selection under secondary user-specific measures (e.g., AUC in classification).
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CV
Enhancing Multimodal Large Language Models for Safety-Critical Driving Video Analysis
Tomaso Trinci, Henrique Piñeiro Monteagudo and Leonardo Taccari
Recent advancements in
Multimodal Large
Language Models (MLLMs) have demonstrated impressive capabilities in general visual understanding. However, their application to safety-critical driving scenarios remains limited by an inability to accurately perceive and reason about rare high-stakes dynamic events, such as collisions or near-collisions. To address this, we introduce a pipeline that enhances MLLM perception by fusing downsampled
video frames with synchronized high-frequency telematics data (IMU and GPS) and semantic insights from specialized computer
vision models. Our pipeline generates high-quality pseudo-labels, including descriptive captions and question-answer pairs, specifically designed to train MLLMs to identify and describe Safety-Critical Events (SCEs) in real-world driving footage. We show the effectiveness of our approach fine-tuning the open-source QwenVL-2.5 model via DoRA adapters: our experiments demonstrate significant improvements in identifying and explaining safety-critical events, with fewer than 50M trainable parameters and limited computational budget.
Read the full abstract ▾
Machine Learning
IKNO: Infinite-order Kernel Neural Operators
Pengyuan Zhu, Ivor W. Tsang and Yueming Lyu
Neural operators have achieved significant success in modern scientific computing due to their flexibility and strong generalization capabilities. Existing models, however, primarily rely on first-order kernel integral approximations, which severely limit their expressivity. To address this, we propose the Infinite-order Kernel Neural Operator (IKNO), which constructs neural operators via infinite-order kernel integrals and admits an elegant closed-form finite approximation. We develop two complementary infinite-order neural operator constructions: IKNO-Vanilla, which applies the full-kernel resolvent on the product grid via Kronecker eigendecomposition, and IKNO-TP, an alternative tensor-product operator that composes per-axis resolvents. Furthermore, we develop fast computation schemes for both variants of IKNO, which achieve outstanding global information aggregation while maintaining high computational efficiency. Empirically, we evaluate our IKNO on both time-dependent and time-independent
benchmarks with arbitrary input shapes, including large-scale industrial
datasets. Extensive experiments demonstrate that the IKNO method consistently achieves the SOTA accuracy with significant improvements on nearly all
benchmark datasets while maintaining scalability to very large point clouds.
Read the full abstract ▾
Machine Learning
Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles
Jinyang Wu, Guocheng Zhai, Ruihan Jin et al. (+7 more)
The proliferation of large
language models (LLMs) and modular skills has endowed autonomous
agents with increasingly powerful capabilities. Existing frameworks typically rely on monolithic LLMs and fixed logic to interface with these skills. This gives rise to a critical bottleneck: different LLMs offer distinct advantages across diverse domains, yet current frameworks fail to exploit the complementary strengths of models and skills, thereby limiting their performance on downstream tasks. In this paper, we present Maestro (
Multimodal Agent for Expert-Skill Targeted Reinforced Orchestration), a Reinforcement Learning (RL)-driven orchestration framework that reframes heterogeneous
multimodal tasks as a sequential decision-making process over a hierarchical model-skill registry. Rather than consolidating all knowledge into a single model, Maestro trains a lightweight policy to dynamically compose ensembles of frozen expert models and a two-tier skill library, deciding at each step whether to invoke an external expert, which model-skill pair to select, and when to terminate. The policy is optimized via outcome-based RL, requiring no step-level supervision. We evaluate Maestro across ten representative
multimodal benchmarks spanning mathematical
reasoning, chart understanding, high-resolution perception, and domain-specific analysis. With only a 4B orchestrator, Maestro achieves an average accuracy of 70.1%, surpassing both GPT-5 (69.3%) and Gemini-2.5-Pro (68.7%). Crucially, the learned coordination policy generalizes to unseen models and skills without retraining: augmenting the registry with out-of-domain experts yields a 59.5% average on four challenging
benchmarks, outperforming all closed-source baselines. Maestro further maintains high computational efficiency with low latency. The source code is available at https://github.com/jinyangwu/Maestro.
Read the full abstract ▾
All authors Jinyang Wu, Guocheng Zhai, Ruihan Jin, Yuhao Shen, Zhengxi Lu, Fan Zhang, Haoran Luo, Zheng Lian, Zhengqi Wen, Jianhua Tao
Machine Learning · Cross-listed from cs.AI
Measuring Cross-Modal Synergy: A Benchmark for VLM Explainability
Joël Roman Ky, Salah Ghamizi and Maxime Cordy
Vision-
Language Models (VLMs) map complex visual inputs to semantic spaces, but interpreting the cross-modal
reasoning of VLMs currently relies on post-hoc explainers evaluated via unimodal perturbation metrics. We expose a limitation in this paradigm: because
multimodal datasets contain
language priors and modality biases, VLMs frequently exhibit cross-modal redundancy, allowing them to answer visual queries using text alone. Consequently, unimodal metrics penalize faithful explainers, triggering an evaluation collapse where visual and textual rankings fundamentally contradict each other. To resolve this, we introduce Synergistic Faithfulness ($\mathcal{F}_{syn}$), a scalable metric rooted in the Shapley Interaction Index that strictly isolates the joint Harsanyi dividend between modalities, serving as a highly accurate surrogate ($\rho = 0.93$) while achieving up to $88\times$ computational speedup. Evaluating 9 distinct XAI methods across 3 VLM architectures and 3
benchmark datasets, reveals that explainers proposed for VLMs heavily over-index on visual salience and significantly underperform adapted
attention-based methods in capturing true cross-modal synergy. By decoupling visual plausibility from cross-modal faithfulness, this work provides a rigorous evaluation framework required to safely audit VLM
reasoning in high-stakes deployments. Code: https://github.com/serval-uni-lu/xai-vlms-
benchmark
Read the full abstract ▾
Machine Learning
World Model Control by Trajectory Reachability Metrics
Liangyu Li, Shengzhi Wang, Libin Qiu et al. (+2 more)
Latent world models can learn representations that contain information needed for control, while the downstream controller may still rank candidate actions poorly when it relies on terminal latent distance alone. We study this failure in a fixed encoder and introduce trajectory reachability metrics (TRM), a small temporal pairwise cost trained from logged trajectories and used to rank predicted endpoints of a candidate action sequence against a goal. In the TwoRoom evaluation on 100 episodes from a high distance range, the original controller with LeWorldModel reaches 7.0% mean success. Temporal TRM trained after excluding all evaluation episodes reaches 96.7%, shuffled label controls stay at 0.0%, and TRM also improves PLDM from 32.7% to 84.0%. TopoNav, a separate pixel
navigation task, shows the same link between repaired ranking and closed loop control. The selection audit with shared candidates (SASC) and rowspace interventions show that the gain comes from reweighting latent directions carrying the action decision. The XY rowspace contributes less than 1% of terminal latent MSE but carries most of the information needed for control. Coverage and boundary tests define the scope. Balanced doorway coverage restores a 0.0% coverage failure to 100.0% success, while an unseen wall orientation and contact rich PushT expose layout, dynamics, and recovery limits.
Read the full abstract ▾
All authors Liangyu Li, Shengzhi Wang, Libin Qiu, Mingliang Xiong, Qingwen Liu
Machine Learning · Cross-listed from astro-ph.IM
Spectra as Language: Large Language Models for Scalable Stellar Parameter and Abundance Inference
Hai-Ling Lu, Yu-Yang Li, Yin-Bi Li et al. (+4 more)
Stellar spectra encode key information on the physical properties and chemical compositions of
stars. Accurate
stellar parameter determination is essential for addressing major questions such as
galaxy and
stellar evolution. Large-scale spectroscopic surveys have accumulated unprecedented spectral data. Traditional feature extraction or model-fitting approaches struggle with high-dimensional, massive
datasets, limited generalization, and computational inefficiency. Recent advances in large
language models demonstrate strong generalization and feature-learning in tasks like natural
language processing, DNA/RNA sequence analysis, and
protein/chemical parsing.
Stellar spectra are continuous sequential signals, enabling the transfer of
language models to
stellar spectroscopy. Here, we propose a two-stage large
language model framework for
stellar parameter inference, achieving accurate estimation of effective temperature, surface
gravity,
metallicity, and abundances of ~20 chemical elements. Scaling-law analyses show systematic performance improvements with increasing data, providing a scalable framework for forthcoming large-scale surveys.
Read the full abstract ▾
All authors Hai-Ling Lu, Yu-Yang Li, Yin-Bi Li, Cun-Shi Wang, A-Li Luo, Jun-Chao Liang, Shuo Li
Machine Learning
One-Way Policy Optimization for Self-Evolving LLMs
Shuo Yang, Jinda Lu, Kexin Huang et al. (+6 more)
Reinforcement Learning with Verifiable Rewards (RLVR) has become a promising paradigm for scaling
reasoning capabilities of Large
Language Models (LLMs). However, the sparsity of binary verifier rewards often leads to low efficiency and optimization instability. To stabilize training, existing methods typically impose
token-level constraints relative to a reference policy. We identify that such constraints penalize deviations indiscriminately; this can flip verifier-determined direction when the policy attempts to outperform the reference, thereby suppressing gains. To resolve this, we propose One-Way Policy Optimization (OWPO), a method based on the principle of decoupling optimization direction from update magnitude. In OWPO, the verifier dictates the update direction, while the reference policy serves only to adjust the magnitude. Specifically, OWPO applies asymmetric reweighting: it performs Accelerated
Alignment for inferior deviations (where the policy lags behind the reference) and Gain Locking for superior deviations (where the policy surpasses the reference). Furthermore, by incorporating iterative reference updates, OWPO creates a “Ratchet Effect” that continuously consolidates gains. Experimental results demonstrate that OWPO outperforms strong baselines, including DAPO, OPD, and MOPD, breaking the bottleneck of fixed priors to enable continuous self-
evolution without reliance on external reference models.
Read the full abstract ▾
All authors Shuo Yang, Jinda Lu, Kexin Huang, Chiyu Ma, Shaohang Wei, Yuyang Liu, Guoyin Wang, Jingren Zhou, Li Yuan
Machine Learning
Algebraic Machine Learning for Small-to-Medium Datasets Is Competitive against Strong Standard Baselines
David Mendez, Fernando Martin-Maroto and Gonzalo G. de Polavieja
Symbolic methods are generally not considered competitive with strong modern learners on realistic supervised tasks. We evaluate Algebraic Machine Learning (AML), a framework that learns through subdirect decomposition of algebraic structure rather than numerical optimization, against standard baselines on image and tabular classification across varying training-set sizes. We find that AML trained only on training data without using validation or cross-validation outperforms a family of cross-validated baseline methods including CNNs on small to medium image
datasets (50–2000 training examples). On tabular
datasets in the same size range, XGBoost is overall the best performing method, but AML is nonetheless comparable to methods incorporating task-specific biases such as LightGBM and random forests. AML achieves this competitive performance across two very different types of
datasets using a generic algebraic inductive bias, rather than the modality-specific biases built into standard baselines like CNNs for images or XGBoost for tabular data, and requires no cross validation because it has no task-dependent hyperparameters to tune.
Read the full abstract ▾
Machine Learning
Short-Term-to-Long-Term Memory Transfer for Knowledge Graphs under Partial Observability
Taewoon Kim, Vincent François-Lavet and Michael Cochez
Reinforcement learning under partial observability requires deciding what information to retain, yet most
memory-based approaches do not explicitly model short-term-to-long-term transfer of symbolic observations. We study this transfer process in a temporal knowledge-graph
memory setting and cast it as a neuro-symbolic value-based decision problem: for each observed triple, the
agent chooses whether to keep or drop it before long-term insertion. To handle variable-sized short-term buffers, we use a per-item Q-learning design with shared parameters and a practical temporal-difference update over matched items across consecutive steps. On the RoomKG
benchmark at long-term
memory capacity 128, learned transfer decisions outperform symbolic and neural baselines, including symbolic baselines with temporal annotations and history-based LSTM/
Transformer baselines. Across transfer-policy ablations, a lightweight local short-term-only variant performs best, and step-level behavior shows that the policy keeps
navigation- and query-relevant facts while discarding lower-value candidate facts, supporting explicit and interpretable
memory decisions under
memory constraints.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.AI
Efficient Agentic Reasoning Through Self-Regulated Simulative Planning
Mingkai Deng, Jinyu Hou, Lara Sá Neves et al. (+4 more)
How should an
agent decide when and how to plan? A dominant approach builds
agents as reactive policies with adaptive computation (e.g., chain-of-thought), trained end-to-end expecting planning to emerge implicitly. Without control over the presence, structure, or horizon of planning, these systems dramatically increase
reasoning length, yielding inefficient
token use without reliable accuracy gains. We argue efficient agentic
reasoning benefits from decomposing decision-making into three systems: simulative
reasoning (System II) grounding deliberation in future-state prediction via a world model; self-regulation (System III) deciding when and how deeply to plan via a learned configurator; and reactive execution (System I) handling fine-grained action. Simulative
reasoning provides unified planning across diverse tasks without per-domain engineering, while self-regulation ensures the planner is invoked only when needed. To test this, we develop SR$^2$AM (Self-Regulated Simulative
Reasoning Agentic LLM), realizing both as distinct stages within an LLM's chain-of-thought, with the LLM as world model. We explore two instantiations: recording decisions from a prompted multi-module system (v0.1) and reconstructing structured plans from traces of pretrained
reasoning LLMs (v1.0), trained via supervised then reinforcement learning (RL). Across math, science, tabular analysis, and web information seeking, v0.1-8B and v1.0-30B achieve Pass@1 competitive with 120-355B and 685B-1T parameter systems respectively, while v1.0-30B uses 25.8-95.3% fewer
reasoning tokens than comparable agentic LLMs. RL increases average planning horizon by 22.8% while planning frequency grows only 2.0%, showing it learns to plan further ahead rather than more often. More broadly, learned self-regulation instantiates a principle we expect to extend beyond planning to how
agents govern their own learning and adaptation.
Read the full abstract ▾
All authors Mingkai Deng, Jinyu Hou, Lara Sá Neves, Varad Pimpalkhute, Taylor W. Killian, Zhengzhong Liu, Eric P. Xing
Machine Learning · Cross-listed from stat.ML
From Betting to Empirical Bernstein LIL
Francesco Orabona
This is a verbatim copy of a technical report I wrote in 2017-2018 to obtain the law of the iterated logarithm using the guarantee on the wealth of an online betting strategy.
Machine Learning · Cross-listed from astro-ph.HE
Self-Supervised ConvLSTM for Fermi Large Area Telescope Transient Detection
Alberto Garinei, Stefano Speziali, Alessandro Vispa et al. (+11 more)
We present a framework for detecting transient gamma-ray phenomena in a controlled environment by combining end-to-end simulations of the
Fermi-LAT sky with self-supervised spatio-temporal deep learning. We generate a ten-year synthetic Universe with gtobssim and process the simulated events into daily all-sky maps of counts and exposure, obtaining a time-ordered sequence that mirrors the structure of
Fermi-LAT observations. To model the nominal
evolution of the sky, we employ a Convolutional Long Short-Term
Memory (ConvLSTM) network that operates directly on map sequences, preserving spatial locality while learning temporal dependencies. The model is trained to reconstruct expected emission, and departures from the learned baseline are quantified through pixel-wise mean-squared residual maps. We then define statistically motivated anomaly criteria by estimating per-pixel thresholds from the residual distribution on the training set, and we enforce spatial coherence via local filtering to suppress isolated fluctuations. The ConvLSTM is then deployed as trained predictor on
Fermi-LAT daily maps, where the sky can depart from the nominal behavior because of genuine astrophysical variability and instrumental non-stationarities. The resulting pipeline flags localized, time-dependent excesses consistent with high-variable sources or transient events (e.g.,
flares or GRBs) and provides a
benchmark for evaluating anomaly-detection strategies on long-duration,
Fermi-LAT-like
datasets.
Read the full abstract ▾
All authors Alberto Garinei, Stefano Speziali, Alessandro Vispa, Andrea Marini, Sara Cutini, Emanuele Piccioni, Marcello Marconi, Francesco Longo, Matteo Martini, Francesca Fallucchi, Romeo Giuliano, Ernesto William De Luca, Umberto Di Matteo, Sabino Meola
Machine Learning
Aerodynamic force reconstruction using physics-informed Gaussian processes
Gledson Rodrigo Tondo, Igor Kavrakov and Guido Morgenthal
Accurate modeling of aerodynamic loads is essential for understanding and predicting the responses of complex structural systems. However, these models often rely on simplifications of the true physical forces, introducing assumptions that can limit their accuracy. Validating such models becomes particularly challenging in the presence of noisy or incomplete data. To address this, we introduce a probabilistic physics-informed machine learning approach designed to reconstruct the underlying aerodynamic loads from noisy measurements of structural dynamic responses. The model avoids overfitting, eliminates the need for regularization schemes, and allows for the use of heterogeneous and multi-fidelity data during the training process. The efficacy of the approach is demonstrated through the reconstruction of aerodynamic loads on the Great Belt East Bridge, simulated under a linear unsteady assumption. Results show a strong agreement between true and predicted loads, particularly related to root mean squared errors, magnitude, phase angle and peak values of the signals. The method for load reconstructing holds broad applicability, such as modeling validation, future load estimation, and structural damage prognosis.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CV
TextTeacher: What Can Language Teach About Images?
Tobias Christian Nauen, Stanislav Frolov, Brian Bernhard Moser et al. (+3 more)
The platonic representation hypothesis suggests that sufficiently large models converge to a shared representation geometry, even across modalities. Motivated by this, we ask: Can the semantic knowledge of a
language model efficiently improve a
vision model? As an answer, we introduce TextTeacher, a simple auxiliary objective that injects text
embeddings as additional information into image classification training. TextTeacher uses readily available image captions, a pre-trained and frozen text encoder, and a lightweight projection to produce semantic anchors that efficiently guide representations during training while leaving the inference-time model unchanged. On ImageNet with standard ViT backbones, TextTeacher improves accuracy by up to +2.7 percentage points (p.p.) and yields consistent transfer gains (on average +1.0 p.p.) under the same recipe and compute. It outperforms
vision knowledge
distillation, yielding more accuracy at a constant compute budget or similar accuracy, but 33% faster. Our analysis indicates that TextTeacher acts as a feature-space preconditioner, shaping deeper layers in the first stages of training, and aiding generalization by supplying complementary semantic cues. TextTeacher adds negligible overhead, requires no costly
multimodal training of the target model and preserves the simplicity and latency of pure
vision models. Project page with code and captions: https://nauen-it.de/publications/text-teacher
Read the full abstract ▾
All authors Tobias Christian Nauen, Stanislav Frolov, Brian Bernhard Moser, Federico Raue, Ahmed Anwar, Andreas Dengel
Machine Learning · Cross-listed from quant-ph
Q-PhotoNAS: Hybrid Quantum Neural Architecture Search Framework on Photonic Devices
Farah Elnakhal, Alberto Marchisio, Nouhaila Innan et al. (+2 more)
Photonic
quantum computing is a promising platform for scalable
quantum machine learning, but designing effective hybrid architectures remains challenging under hardware and optimization constraints. Existing approaches rely on manually tuned architectures that fail to account for the collaboration between classical preprocessing, phase encoding, and photonic
circuit structure, limiting both accuracy and hardware compatibility. In this paper, we propose a neural architecture search framework for hybrid photonic
quantum-classical models that combines genetic
algorithm-based search with learnable
quantum phase encoding to systematically explore the joint design space of classical and
quantum components. Our framework encodes 19 hyperparameters across six gene groups and evolves a population of hybrid architectures using group-based crossover, per-gene mutation, and elitism, evaluating each candidate on a short training budget before full retraining of the best found design. We evaluate our framework on two image classification
benchmarks, Digits and MNIST, achieving final validation accuracies of 99.44% and 98.78%, respectively, with first-principles execution time estimates on the Quandela Ascella photonic QPU projecting single-image inference at ~67 ms (Digits) and ~149 ms (MNIST). Our
quantum contribution analysis further shows that the photonic layer extracts non-redundant features orthogonal to the classical pathway, providing a measurable accuracy advantage over classical-only baselines. Our results demonstrate that automated architecture search is both practical and impactful for hybrid photonic systems, opening the way for systematic design space exploration of
quantum AI on photonic devices.
Read the full abstract ▾
All authors Farah Elnakhal, Alberto Marchisio, Nouhaila Innan, Gabriel Falcao, Muhammad Shafique
Machine Learning · Cross-listed from cs.SD
RobustSpeechFlow: Learning Robust Text-to-Speech Trajectories via Augmentation-based Contrastive Flow Matching
Jinhyeok Yang, Hyeongju Kim, Yechan Yu et al. (+3 more)
While flow-matching text-to-
speech (TTS) achieves strong zero-shot speaker similarity and naturalness, it remains susceptible to content fidelity issues, particularly skip and repeat errors from imperfect
alignment. We propose RobustSpeechFlow, a training strategy that improves
alignment robustness by extending contrastive flow matching with length-preserving repeat and skip latent augmentations. Requiring no external aligners or preference data, our method directly penalizes realistic failure modes and readily integrates into existing pipelines. On Seed-TTS-eval, it reduces the word error rate (WER) from 1.44 to 1.38 using only 0.06B parameters. On our ZERO500
benchmark, it delivers consistent intelligibility improvements across diverse speaker and prosody conditions; at NFE=24, it reduces English character error rate (CER) from 0.48% to 0.35% and Korean CER from 0.81% to 0.57%. Audio samples: https://robustspeechflow.github.io/
Read the full abstract ▾
All authors Jinhyeok Yang, Hyeongju Kim, Yechan Yu, Joon Byun, Frederik Bous, Juheon Lee
Machine Learning · Cross-listed from cs.RO
CoRMA: Contrastive RMA for Contact-Rich Meta-Adaptation
Wentian Wang, Chutong Wen, Hongxu Ma et al. (+6 more)
We present CoRMA(Contrastive Robotic Motor Adaptation), a context-based meta-adaptation framework that modifies RMA for force-dominant assembly. CoRMA replaces raw simulator-parameter adaptation with a compact 6D simulator-only semantic contact context describing contact onset, lateral engagement, guided transition, contact direction, and jamming. A deployable causal
Transformer adapter infers this context online from force, proprioceptive, and action histories using semantic regression and a force-regime contrastive objective. At deployment, oracle context is removed and replaced by the inferred context, enabling within-episode adaptation without demonstrations, privileged inputs, or gradient updates. We evaluate CoRMA on PegInsert, GearMesh, and NutThread in Isaac Lab / Isaac Sim 5.0 and on a real Marvin arm. Compared with FORGE baselines that achieve high simulation success but degrade substantially on hardware, CoRMA retains higher verified real success under controlled target-pose noise. These results support semantic contact inference as a reusable adaptation interface within a related assembly task family, while broader unseen-task generalization and Real2Sim calibration remain future work.
Read the full abstract ▾
All authors Wentian Wang, Chutong Wen, Hongxu Ma, Wuhao Wang, Zhexiong Xue, Abdul Haseeb Nizamani, Dandi Zhou, Xinhai Sun, Jianqiao Zhu
Machine Learning
Can Breath Biomarkers Causally Influence Blood Glucose? Investigating VOC-Mediated Modulation in Diabetes
Varsha Sharma, Prasanta K. Guha and Avik Ghose
Diabetes is a global health burden, and early detection is critical for timely intervention. This study explores a non-invasive, data-driven framework to identify individuals at risk of diabetes using Volatile Organic Compounds (VOCs) and lifestyle variables. We use causal inference techniques to estimate the impact of VOCs such as acetone, isopropanol, isoprene, and ethanol on blood glucose levels. Additionally, we designed a classifier to distinguish diabetics from non-diabetics using non-invasive markers. We created a risk-based ranking system for individuals in the "gray zone," and identified natural clusters in the population using Gaussian Mixture Model. Our results suggest that specific VOCs exhibit a strong causal influence on glucose levels and that machine learning models can reliably classify and stratify individuals at high risk. This integrated causal-explainable analysis can support the development of tool for non-invasive early screening of diabetes.
Read the full abstract ▾
Machine Learning
From Reasoning Chains to Verifiable Subproblems: Curriculum Reinforcement Learning Enables Credit Assignment for LLM Reasoning
Xitai Jiang, Zihan Tang, Wenze Lin et al. (+3 more)
Reinforcement learning from verifiable rewards (RLVR) has shown strong promise for LLM
reasoning, but outcome-based RLVR remains inefficient on hard problems because correct final-answer rollouts are rare and sample-level credit assignment cannot use partial progress in failed attempts. We introduce SCRL (Subproblem Curriculum Reinforcement Learning), a curriculum RL framework that derives verifiable subproblems from reference
reasoning chains and fixes the final subproblem as the original problem. This turns partial progress on hard problems into verifiable learning signals. Algorithmically, SCRL uses subproblem-level normalization, which normalizes rewards independently at each subproblem position and assigns the resulting advantages to the corresponding answer spans, enabling finer-grained credit assignment without external rubrics or reward models. Our analysis shows that subproblem curricula lift hard problems out of gradient dead zones, with larger relative gains as the original problem becomes harder. Across seven mathematical
reasoning benchmarks, SCRL outperforms strong curriculum-learning baselines, improving average accuracy over GRPO by +4.1 points on Qwen3-4B-Base and +1.9 points on Qwen3-14B-Base. On AIME24, AIME25, and IMO-Bench, SCRL further improves pass@1 by +3.7 points and pass@64 by +4.6 points on Qwen3-4B-Base, indicating better exploration on hard
reasoning problems.
Read the full abstract ▾
All authors Xitai Jiang, Zihan Tang, Wenze Lin, Yang Yue, Shenzhi Wang, Gao Huang
Machine Learning · Cross-listed from cs.CV
TWINGS: Thin Plate Splines Warp-aligned Initialization for Sparse-View Gaussian Splatting
Hyeseong Kim, Geonhui Son, Deukhee Lee et al. (+1 more)
Novel view synthesis from sparse-view inputs poses a significant challenge in 3D computer
vision, particularly for achieving high-quality scene reconstructions with limited viewpoints. We introduce TWINGS, a framework that enhances 3D Gaussian Splatting (3DGS) by directly addressing point sparsity. We employ Thin Plate Splines (TPS), a smooth non-rigid deformation model that minimizes bending energy to estimate a globally coherent warp from control-point correspondences, to align backprojected points from estimated depth with triangulated 3D control points, yielding calibrated backprojected points. By sampling these calibrated points near the control points, TWINGS provides a fast and geometrically accurate initialization for 3DGS, ultimately improving structural detail preservation and color fidelity in reconstructed scenes. Extensive experiments on DTU, LLFF, and Mip-NeRF360 demonstrate that TWINGS consistently outperforms existing methods, delivering detailed and accurate reconstructions under sparse-view scenarios.
Read the full abstract ▾
All authors Hyeseong Kim, Geonhui Son, Deukhee Lee, Dosik Hwang
Machine Learning
Prototype-Guided Classification Sub-Task Decoupling Framework: Enhancing Generalization and Interpretability for Multivariate Time Series
Xianhao Song, Yuang Zhang, Yuqi She et al. (+2 more)
Time Series Classification (TSC) is a long-standing research problem that has gained increasing
attention in recent years with the rapid growth of large-scale temporal data. Despite substantial progress enabled by deep learning, designing TSC models that are both accurate and interpretable remains a challenging task. Many existing approaches adopt a direct feature-to-label classification paradigm, by collapsing high-dimensional temporal
embeddings into class logits via a single linear projection (often after global pooling), the paradigm conflates feature extraction and decision logic into an inseparable mapping. To address these limitations, we propose PDFTime, a prototype-guided framework that reformulates time series classification as a multi-stage decision process. Instead of direct feature-to-label mapping, PDFTime leverages learned prototypes to approximate class-conditional feature distributions in the latent space, enabling progressive discrimination through classification sub-tasks of varying granularity. To our knowledge, PDFTime is the first framework to reformulate time series classification as a decoupled, multi-stage similarity-based
reasoning process, breaking the long-standing paradigm of direct, black-box feature-to-label mapping. Extensive evaluations demonstrate that PDFTime achieves state-of-the-art (SOTA) performance across UEA and UCR
benchmarks. Notably, it secures the top-$1$ accuracy on 80 out of 128
datasets in the UCR archive, significantly outperforming recent strong baselines in both consistency and generalization.
Read the full abstract ▾
All authors Xianhao Song, Yuang Zhang, Yuqi She, Liping Wang, Xuemin Lin
Machine Learning
LABO: LLM-Accelerated Bayesian Optimization through Broad Exploration and Selective Experimentation
Zhuo Chen, Xinzhe Yuan, Jianshu Zhang et al. (+8 more)
The high cost and data scarcity in scientific exploration have motivated the use of large
language models (LLMs) as knowledge-driven components in
Bayesian optimization (BO). However, existing approaches typically embed LLMs directly into the sampling or surrogate modeling pipeline, without fully leveraging their significantly lower evaluation cost compared to real-world experiments. To address this limitation, we propose LLM-Accelerated
Bayesian Optimization (LABO), a framework that combines LLM predictions with experimental observations within a single BO loop. LABO employs a gating criterion to dynamically balance the reliance on LLM predictions versus actual experiments. By leveraging inexpensive LLM evaluations to broadly explore the search space and reserving costly real experiments only for regions with high uncertainty, LABO achieves more sample-efficient optimization. We provide a theoretical analysis with a cumulative regret bound that formalizes this efficiency gain. Empirical results across diverse scientific tasks demonstrate that LABO consistently outperforms existing methods under identical experimental budgets. Our results suggest that LABO offers a practical and theoretically grounded approach for integrating LLMs into scientific discovery workflows.
Read the full abstract ▾
All authors Zhuo Chen, Xinzhe Yuan, Jianshu Zhang, Jinzong Dong, Ruichen Zhou, Yingchun Niu, Tianhang Zhou, Yu Yang Fredrik Liu, Yuqiang Li, Nanyang Ye, Qinying Gu