Optimization
-
Automatic Differentiation with Constraints in Forward and Inverse Problems
2026-10-20 09:11:38Nikolay Yavich, Alexander Ryabov, Viacheslav Naumov, Sayan Ranu, N. M. Anoop Krishnan, Evgeny Burnaev, Vladimir VanovskiyAbstract
Approximate solutions of partial differential equations (PDEs) on unstructured grids can be represented on a graph. Numerical optimization involving stiff evolutionary and static PDEs often requires solving a large system of equations, which can be treated as a constraint. In this work, we present a fully differentiable graph-based implicit solver that incorporates this constraint. We evaluate its performance across four key tasks: a) linear and nonlinear forward modeling of evolutionary and static problems, b) solving inverse problems to estimate equation coefficients, c) solving inverse problems to estimate the right-hand side, and d) learnable graph coarsening to accelerate forward modeling. The stability and differentiability of our solver enable significant reductions in the complexity of forward modeling while efficiently addressing related inverse problems. These capabilities make it a powerful tool for different physics and engineering applications, in particular for geophysics, where inverse problems are commonly solved in order to reduce the model uncertainty.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27888
PDF
-
From Generic to Dedicated: A Novel Optimizer for Online Continual Learning
Abstract
Online Continual Learning (OCL) requires models to learn from non-stationary data streams under a strict single-pass constraint, making them highly susceptible to catastrophic forgetting. While existing studies have explored various strategies from data, architecture, and optimization perspectives, the optimizer often directly uses generic approaches such as SGD and AdamW. In this work, we reveal a new link between optimization dynamics and OCL needs by recasting Polyak-Ruppert averaging as the engine of "plasticity" and Primal averaging as the anchor of "stability". This inspires our novel optimizer, SPIN (Stability-Plasticity INterpolation), which explicitly decouples plasticity and stability by interpolating between a fast-moving "plasticity" sequence and an adaptive "stability" sequence. We theoretically demonstrate that SPIN implicitly employs an inverse Hessian approximation, providing crucial Tikhonov regularization to damp noisy, single-sample Hessian estimates and mitigate forgetting. More importantly, SPIN can be seamlessly integrated into existing OCL methods by taking resulted gradients as input and replacing standard optimizers. Extensive experiments show that using SPIN instead of standard optimizers yields consistent and significant performance gains across multiple OCL approaches, including the challenging rehearsal-free setting and under strict memory constraints.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27835
PDF
-
Geometry Regularization for Vanilla NeRF via Self-Supervised Cross-Ray Consistency
2026-10-20 07:40:14Zongji WangAbstract
Neural Radiance Fields (NeRF) achieve high-quality novel-view synthesis by representing a scene as a continuous volumetric radiance field. However, the geometry encoded by the density field of vanilla NeRF can be noisy and inaccurate, since different combinations of shape and radiance may explain the same input images. To reduce this ambiguity, we augment vanilla NeRF with a geometry-optimization branch based on self-supervised cross-ray consistency. Specifically, this branch samples candidate surface points from the density field, generates virtual rays that intersect at each point from different directions, and renders their appearance using the NeRF network. Based on the visibility-weighted color consistency across these rays, we constrain the implicit geometry. We further regularize the implicit geometry by penalizing opacity at candidates with inconsistent appearance predictions and encouraging local normal smoothness. These geometric constraints are derived from the NeRF predictions and require no ground-truth depth or surface-normal supervision. Experiments on eight Replica scenes show that the proposed method reduces the mean Chamfer distance by 30.8% compared with vanilla NeRF.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27776
PDF
-
How Much Do $ell_1$ Attacks Need Concentrated Perturbations?
2026-10-20 07:37:48Melika Shirian, Kianoosh Vadaei, Arshia HemmatAbstract
A global $\ell_1$ budget limits how much an adversary may change an image, not how that change is distributed. We ask how much of an $\ell_1$ attack's observed success remains when large coordinate changes are restricted. We add a per-coordinate cap to the budget, enforce the joint constraint by exact in-loop projection, and sweep the cap for APGD-$\ell_1$, SLIDE, and FMN-$\ell_1$ without altering their objectives. Across 18 standard-model settings, caps of at most $32/255$ retain a median $97\%$ of attack success relative to each method's native uncapped baseline, while median peak magnitude falls by $80\%$ and median effective support nearly triples. At matched realized $\ell_1$, capped perturbations remain structurally distinct; separate ImageNet comparisons also show lower LPIPS and DISTS at selected caps. Direct in-loop constrained optimization preserves up to $88$ percentage points more attack success than post-hoc clipping, highlighting the importance of constraint enforcement. Suitable caps nevertheless vary across attacks and models, and adversarially trained classifiers exhibit more heterogeneous concentration sensitivity. These results establish perturbation concentration as a complementary, model- and attack-dependent axis of $\ell_1$ attack evaluation.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27773
PDF
-
PixelSR: Pixel-Space Orthogonal Flow Decoupling for Real-World Image Super-Resolution
2026-10-20 07:21:29Xiang Lv, Kai Zhao, Qiang Wang, Qiao Mo, Chao Zhou, Mingwen ShaoAbstract
Latent diffusion models have recently achieved remarkable progress in Real-world Image Super-Resolution (Real-SR). However, existing methods consistently exhibit a persistent seesaw effect: enhancing perceptual realism compromises structural fidelity. This effect, we argue, is not intrinsic to diffusion models, but largely stems from the architectural constraints of prevailing two-stage frameworks. To address this challenge, we propose PixelSR, a novel single-stage Pixel Diffusion Transformer for Real-SR that operates directly in pixel space. Specifically, PixelSR consists of two core designs: Pixel-space Conditional Diffusion and Orthogonal Flow Decoupling. The former establishes an effective pixel-space restoration framework by integrating spatially aligned LR conditioning, visual semantic guidance, and lightweight texture refinement, thereby bypassing VAE compression and enabling end-to-end optimization directly in RGB space. Complementarily, the latter decomposes the asymmetric flow target into a low-rank semantic component and a noise-free orthogonal texture component. By routing these components to specialized branches, it achieves decoupled optimization with reduced mutual interference. Benefiting from these designs, our PixelSR restores perceptually realistic details while faithfully preserving structural content. Extensive experiments across multiple benchmarks demonstrate that PixelSR not only achieves competitive performance, but also effectively mitigates the fidelity--realism seesaw effect.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27759
PDF
-
APOLO: Automatic Prompt Optimization for Ontology Learning
2026-10-20 07:08:17Huu Tan Mai, Roman Kochnev, Cuong Xuan Chu, Lukas Lange, Heiko Paulheim, Daria StepanovaAbstract
Ontology Learning (OL) from text has advanced with the emergence of Large Language Models (LLMs), but it remains challenging due to the limited availability of annotated training data and the difficulty of adapting LLMs to perform OL effectively. We address this via APOLO - Automatic Prompt Optimization for Ontology Learning, by casting OL as an explicit prompt optimization problem over LLM modules. To obtain training data, we employ a multi-agent system that generates text-ontology pairs from existing expert-curated ontologies. We then propose two ontology learner architectures: a greedy and an autoregressive learner, and optimize both using GEPA, a greedy evolutionary prompt optimizer built on DSPy. Experiments on two ontologies - a biomedical (DOID) and a plant ontology (PO) show consistent improvements after optimization across nearly all model and mode combinations, with autoregressive learners achieving the largest gains. Our results demonstrate that prompt optimization is a viable and lightweight alternative to fine-tuning for OL, and that the autoregressive formulation better captures ontological structure than the greedy approach.
Abstract (translated)
URL
https://arxiv.org/abs/2609.34540
PDF
-
Prescribed Performance Guided Safe Reinforcement Learning for Roundabout Navigation of Autonomous Vehicles
2026-10-20 06:54:57Wenhan Li, Yangrui Zhang, Hang Zhou, Peiran Liu, Yiding JiAbstract
Connected and Autonomous Vehicles (CAVs) face significant challenges in roundabout navigation due to dense traffic interactions and unpredictable agent behaviors, necessitating high levels of adaptability and safety. While the combination of Reinforcement Learning (RL) and Model Predictive Control (MPC) offers a promising approach, existing frameworks suffer from a structural mismatch where pointwise RL decisions fail to align with MPC's trajectory level optimization, and often lack rigorous safety guarantees. This work proposes RLTrajGen-PPMPC-CBF, a hierarchical framework for safe, smooth and comfort roundabout navigation, which regulates the coupling between RL-based planning and MPC-based control. We employ Prescribed Performance Constraints (PPC) to enforce a gradual, stable convergence between reference trajectories and control outputs, ensuring MPC tracking performance remains consistent with intended behavior from RL. Additionally, a Control Barrier Function (CBF) layer provides real-time safety corrections that are fed back into the RL process. Simulation results in the MetaDrive environment demonstrate that our method achieves a 6.25\% increase in average success rate and a 65.87\% reduction in Motion Sickness Dose Value (MSDV) compared to the optimal baseline, meanwhile accelerating training convergence.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27732
PDF
-
World2HOSI: A Multi-Expert Harness for Long-Horizon Human-Object-Scene Interaction Generation
2026-10-20 06:43:51Jiushu Li, Haoran Wang, Yumeng Liu, Zhiyang Dou, Yuan Liu, Yuexin Ma, Cheng LinAbstract
Long-horizon human-object-scene interaction (HOSI) is central to embodied AI, animation, and game development. Existing human-object interaction (HOI) and human-scene interaction (HSI) models excel in their respective domains, but none alone covers tasks requiring multiple interaction skills. Training unified HOSI models is hindered by scarce and costly data. We introduce World2HOSI, a multi-expert orchestration harness that generates continuous HOSI data from 3D Gaussian Splatting (3DGS) scenes and natural language instructions. Our key insight is that long-horizon HOSI hinges not on a stronger single model but on orchestration. Given a shared, structured scene state and an orchestration harness grounded in it, existing experts can jointly accomplish tasks beyond the reach of any of them alone. World2HOSI realizes this insight as a closed loop around a persistent scene state built from decoupled background and object assets. A harness plans action stages over this state, invokes experts via their native conditioning interfaces, and writes each action's effects back, so that every invocation is grounded in the evolving world. To reduce physical artifacts, bounded residual optimization and contact-driven physics simulation further ensure the physical plausibility of human motion and object manipulation, respectively, while preserving interaction intent. Experiments and a perceptual user study show that World2HOSI achieves the best task completion and interaction quality on the evaluated HOSI tasks, while generating more diverse data.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27725
PDF
-
A Principled Optimal Transport Framework for Frame Selection in Long Video Understanding
2026-10-20 06:12:30Ziwei Zhao, Yuhang LiuAbstract
Long-video understanding with multimodal LLMs is fundamentally constrained by the mismatch between video length and the model's limited visual context budget, making frame selection essential for practical reasoning. Existing query-aware methods typically rely on heuristic pipelines: they assign one-dimensional query-conditioned scores to frames and then apply sampling, ranking, or extraction rules to form the final subset. While effective in practice, such methods are generally not derived from an explicit optimization objective and struggle to capture the heterogeneous evidence required for complex reasoning. In this paper, we revisit long-video frame selection from a principled optimization perspective. We formulate it as a structured evidence-allocation problem and propose OTFS, an optimal transport framework that maps multiple evidence sources onto video frames. To reflect the asymmetry of this problem---important evidence should be preserved, whereas redundant frames need not absorb mass---we instantiate this view with a semi-unbalanced entropic optimal transport objective and an efficient solver, OTFS-Sinkhorn. Combined with an exact dynamic program for coverage-aware subset extraction, OTFS yields a practical two-stage training-free pipeline. We further show that, in the single-source case, our formulation induces a Gibbs-type distribution over frames, making Q-Frame's temperature-scaled distribution a special case of our framework while also providing a broader perspective for understanding related prior methods. Experiments on three long-video understanding benchmarks show that OTFS consistently outperforms strong training-free baselines. Codes are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27692
PDF
-
HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs
2026-10-20 06:00:54Zhuoshi Pan, ZhiYuan Feng, Yukuo Cen, Honglin Lin, Junru Lu, Zhongxia Chen, Shuo Yang, H. Vicky Zhao, Jie TangAbstract
Harness development plays a crucial role in building capable AI agents. Recent approaches improve harnesses through recursive self-improvement (RSI), but their performance gains often come with higher execution costs. We introduce HarnessPro, a label-free method for evolving harnesses to improve task performance while reducing execution cost. In this setting, candidate harnesses are evaluated using LLM-generated tests without access to ground-truth solutions or official tests. Specialized agents diagnose execution trajectories, plan updates with different performance-cost priorities, and implement them as candidate harnesses. Rather than returning a single harness, HarnessPro uses normalized weighted squared-distance selection to provide cost-focused (lite), balanced (medium), and performance-focused (heavy) harnesses for different deployment budgets. Selected harnesses enter the next iteration, where execution feedback guides further updates. Our experiments cover Terminal-Bench 2, SWE-Bench Pro, and ProgramBench, using GPT-5.5 with Codex and GLM-5.2 with Claude Code. Across these benchmarks using GPT-5.5 with Codex, the heavy mode of HarnessPro achieves an average pass-rate gain of 2.0 percentage points and an average per-task cost reduction of 26.1% relative to RHO. This advantage also holds when using GLM-5.2 with Claude Code. Code is available at this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.26770
PDF
-
PACE-MoE: Adaptive Domain Sampling and Domain-Specific LoRA for Medical Visual Grounding
2026-10-20 05:13:55Zhenxuan Wang, Jisheng Dang, Ronghao Lin, Bimei Wang, Jingwen Zhao, Wenjun Ma, Bin Hu, Qi Tian, Tat-Seng ChuaAbstract
Medical visual grounding requires models to align clinical descriptions with spatial evidence across heterogeneous imaging domains. However, existing multi-domain post-training strategies typically rely on fixed domain sampling and shared parameter updates, overlooking the need to adapt both which domains receive training and where their updates are applied. To address this issue, we propose PACE-MoE, a policy-guided framework that jointly performs adaptive domain sampling and domain-specific adaptation for medical visual grounding. PACE dynamically assesses the training value of each imaging domain from current-policy feedback and progressively adjusts domain coverage through domain admission, replay, and final all-domain balancing. Meanwhile, domain-specific LoRA adapters route updates to the corresponding imaging domains while preserving the pretrained backbone and visual encoder. This coordinated design aligns training-data allocation with parameter adaptation without modifying the original reward formulation or Group Relative Policy Optimization (GRPO) objective. Experiments across seven medical imaging domains show that PACE-MoE consistently improves grounding performance over fixed-domain training, achieving a 3.64% relative improvement in mean box IoU over the Vanilla baseline. Under matched domain-specific adapter architectures, adaptive sampling alone further improves mean box IoU by 2.07%, demonstrating the benefit of policy-guided domain allocation. These results establish coordinated data-parameter routing as an effective strategy for multi-domain medical visual grounding.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27656
PDF
-
Learning Goal-Conditioned Subgoals through Frozen Hierarchical World Models
2026-10-20 04:34:51Xiang Kaibo, Liu QingfengAbstract
Hierarchical world models plan over long horizons by passing predicted states from a coarse model to a lower-level controller. Choosing these states can be expensive: the upper-level planner repeatedly differentiates through candidate macro-action trajectories. We replace this search with a lightweight goal-conditioned network trained through the frozen high-level predictor. The network supplies a conditioning vector, while the predictor generates the next latent waypoint. Training uses hindsight goals and state-prediction losses on offline trajectories, without search-solution targets or macro-action labels. Inference retains an autoregressive high-level rollout, but requires no high-level candidate optimization. We evaluate this approach on the shared-latent HWM checkpoints distributed with H-JEPA, using three backbone seeds and three head initializations per backbone. With a common evaluation protocol, mean success is 70.4% on Cube, 48.7% on Push-T, 100.0% on FourRoom, and 71.1% on Visual AntMaze. The method improves over the corresponding HWM search baseline on three environments and has the same mean on Push-T. Dedicated measurements on an NVIDIA L40S give 1.28-4.39x end-to-end speedups for batched evaluation. Ablations identify state-adaptive conditioning as a consistent contributor, while predictor-mediated subgoals and unrestricted conditions have environment-dependent benefits.
Abstract (translated)
URL
https://arxiv.org/abs/2610.26179
PDF
-
Beyond Experience: Knowledge-Guided Evolution for NPU Kernel Generation
2026-10-20 03:56:26Xi Chen, Ming Shen, Haotian Yan, Ruoyu Sun, Tianjun YaoAbstract
LLM agents have become capable kernel engineers on mainstream GPU stacks, but they struggle on emerging accelerators such as Ascend NPUs, whose programming model is barely represented in pretraining data. Recent NPU agents address this gap largely through experience: they accumulate memory from their own generation and optimization trials and reuse it on later tasks. We argue that experience alone is not enough for kernel generation. An agent can only remember what it has tried, and it rarely tries a mechanism it does not know. As a result, such memory offers little help when a new operator calls for a different mechanism, and even familiar operators raise unseen situations as optimization proceeds. We present ASCENDEVOLVE, which grounds every step of an evolutionary kernel search in a large external knowledge base built from platform documentation, toolkit headers, and operator repositories. Because this corpus far exceeds any context window, ASCENDEVOLVE indexes it and routes each agent to a small, task-specific slice of evidence instead of the corpus itself. The retrieved knowledge seeds diverse implementation strategies, guides repair when a candidate fails, and constrains how strong candidates are recombined. On 53 Ascend C operators from CANN-Bench spanning four difficulty levels, ASCENDEVOLVE achieves an overall score of 71.39, improving over Reflexion and OpenEvolve by 11.6% and 17.2%.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27588
PDF
-
Policy Search in Code Space for Multi-Agent Pathfinding
2026-10-20 03:42:05Matthias Eder, Pantelis Emmanouil, Vivek Farias, Fabian Fischer, Adam Jozefiak, Tianyi Peng, Thomas Ulz, Raphael Watschinger, Fangyu Wu, Andrew ZhengAbstract
Developing coordination policies for complex multi-agent systems relies on either labor-intensive algorithm design or opaque task-specific reinforcement learning. With recent advances in autonomous coding agents, we show that policy search directly in code space offers a compelling third path that combines automated optimization with explicit interpretability. We investigate a human-directed workflow in which coding agents iteratively propose, implement, evaluate, diagnose, and eventually refine algorithms guided by simulation feedback and explicit correctness verification. We conducted two case studies and found that this workflow robustly outperformed any manually engineered solutions. On a challenging map in the 2024 League of Robot Runners competition, a two-day search discovered a policy that exceeds the stronger locally reproduced competition entry by 35.0% on the densest instance and generalizes to unseen map instances. In simulated real-world warehouses with irregular lanes, continuous robot geometry, and asynchronous motion, a week-long search discovered an entirely new class of multi-agent pathfinding algorithm that outperforms a calibrated proprietary solution in throughput by 58.9%. Under stated assumptions, we further prove that the new algorithm is collision-free and completes, in finite time, every accepted goal that is not withdrawn or replaced. Together, the studies show that policy search in code space powered by state-of-the-art coding agents offers a highly competitive path to building real engineering solutions that are both effective and analytically tractable.
Abstract (translated)
URL
https://arxiv.org/abs/2610.25946
PDF
-
TRCM: Topology-Aware Route-Consistency Modeling for Robust Neural Vehicle Routing
2026-10-20 03:32:20Peiheng Jian, Zhenwei Wang, Tiehua Zhang, Lei Zhang, Zhishu Shen, Ruibin Bai, Zhijun DingAbstract
Neural combinatorial optimization has enabled fast constructive solvers for the Capacitated Vehicle Routing Problem (CVRP), yet most autoregressive models still construct solutions through sequential node selection and represent route organization only implicitly. We investigate whether route-level customer co-assignment can be modeled explicitly and used to guide this node-level decision process. To this end, we propose TopologyAware RouteConsistency Modeling (TRCM), which reuses node representations from a standard autoregressive construction pipeline to predict pairwise same-route probabilities and injects them into customer selection as a route-consistency signal. Because sameroute supervision is both class-imbalanced and structurally uneven, TRCM combines structure-preserving augmentation with Topological Entropy (TE)- based reweighting to emphasize locally ambiguous customer pairs. Under matched decoding budgets, TRCM reduces the optimality gap from 3.67% to 3.28% on CVRP50 and from 2.11% to 1.99% on CVRP100; under 8-fold augmented inference, it reduces the CVRP100 gap from 1.61% to 1.49%. On XML100, TRCM*8Aug improves the average objective by 0.24%, with larger gains on several structurally informative factors. A controlled demand-reassignment test over 500 variants further shows more concentrated objective fluctuations. These results support explicit route-consistency modeling as an effective structural principle for autoregressive neural routing.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27568
PDF
-
Velocity Optimization via Thermal Imaging Enhances Autonomous Tumor Resection
2026-10-20 03:22:13Naveed D. Riaziat, Jiawei Ge, Ethan Kilmer, Michael W. Mikula, Paul Maria Scheikl, Michael Kam, Leila J. Mady, Axel Krieger, Jeremy D. BrownAbstract
Objective: To demonstrate and evaluate THERMO, a thermal-velocity optimization algorithm for autonomous electrosurgery, in ex vivo porcine tissue using the UR10e-based ASTR system. Methods: Thermography for Electrosurgical Rate Modulation via Optimization (ThERMO), a closed-loop algorithm that uses real-time thermal imaging data to minimize thermal injury while ensuring accurate cut margins, was tested on ex vivo porcine tissue. Histological analysis was performed to detect thermal injury extent at a cellular level, and 3D photogrammetric measurement was used to verify margin accuracy. ThERMO's performance was compared against two previously used fixed velocity methods. Results: ThERMO outperformed both fixed velocity comparison methods without requiring data-driven training or iterative tuning, reducing thermal injury by up to 52%, improving cut accuracy by up to 59%, and maintaining task completion speed. Conclusions: Closed-loop thermal feedback via ThERMO improves autonomous robotic electrosurgery by simultaneously reducing thermal injury and improving margin accuracy, reinforcing its value over fixed velocity approaches for tumor resection applications. Significance: This work advances autonomous robotic tumor resection by demonstrating that real-time closed-loop thermal feedback control can substantially reduce collateral tissue damage and improve surgical margin accuracy.
Abstract (translated)
URL
https://arxiv.org/abs/2610.27562
PDF
-
Secure Agentic Coding through Counterexample-Grounded Feedback
2026-10-20 03:21:13Tue Le, Anh Nguyen Hoang, Gabriel Sasseville, Doseok Jang, Pierre-Luc Bacon, Nghi D. Q. BuiAbstract
As coding models and agent harnesses become more capable, reliable security verification is becoming a central bottleneck. Improving secure code generation requires a signal whose reported progress remains aligned with actual program behavior under optimization; otherwise, an optimizer can improve the signal's verdict without removing the underlying vulnerability. Existing model-based approaches require models to either infer program behavior from source code or specify correct outcomes for adversarial tests, placing substantial reasoning demands on the security signal. We introduce \falcon{}, a counterexample-grounded verification framework that separates attack exploration from behavioral verification and connects them through execution traces of the program under test. In test-time repair, \falcon{} yields substantially stronger held-out functional-security gains than alternative security signals across secure code generation benchmarks, while preserving or improving functionality. As an RL reward, \falcon{} also outperforms static-analysis and learned rewards on held-out functional-security, even when all policies successfully optimize their respective training rewards. Combining \falcon{} with either proxy reward further improves held-out performance, suggesting that execution-grounded feedback can mitigate reward overoptimization while preserving useful guidance from existing security signals.
Abstract (translated)
URL
https://arxiv.org/abs/2610.25916
PDF
-
Expected Return Causes Outcome-Level Mode Collapse in Reinforcement Learning and How to Fix It with Inverse Probability Scaling
2026-10-20 03:12:50Abhijeet Sinha, Sundari Elango, Dianbo LiuAbstract
Many reinforcement learning (RL) problems admit multiple terminal solutions of comparable quality, where the goal is not to identify a single optimum but to represent a diverse set of high-quality outcomes. Nevertheless, policies trained by standard expected return maximization routinely collapse onto a small subset of outcomes, a phenomenon commonly attributed to insufficient exploration or weak regularization. We show that this explanation is incomplete: outcome level mode collapse is a structural consequence of the expected-return objective itself. Under idealized learning dynamics, the log-probability ratio between any two outcomes evolves linearly in their reward difference, implying exponential ratio divergence and inevitable collapse independent of the exploration strategy, entropy regularization, or optimization algorithm. We identify the source of this pathology as the probability multiplier inside the expectation and propose a minimal correction: inverse probability scaling, which removes outcome-frequency amplification from the learning signal, fundamentally changes the learning dynamics, and provably yields reward-proportional terminal distributions, preventing collapse in multimodal settings. We instantiate this principle in Group Relative Policy Optimization (GRPO) as a drop-in modification, IPS-GRPO, requiring no auxiliary models or architectural changes. Across different reasoning and molecular generation tasks, IPS-GRPO consistently reduces outcome-level mode collapse while matching or exceeding baseline performance, suggesting that correcting the objective rather than adding exploration heuristics is key to reliable multimodal policy optimization.
Abstract (translated)
URL
https://arxiv.org/abs/2601.21669
PDF
-
TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
2026-10-20 02:09:40Claudio Basile, Michael Voznesensky, Gunhyun Park, Ben Wilson, Parker Holloway, Zhanyong Wan, Vlad Belous, Manish Khanna, Paul Ganssle, Kevin Gleason, Andrew Dame, Brendan Folie, Bhavya Bahl, Haifeng Jin, Pedro Goncalves Mokarzel, Arian Arfaian, Luke Boyer, Chi Zhang, Yiling Yuan, Adam Mainz, Aditya Patel, Huangze Tang, Jay Gan, Lizzie Liang, Will Froom, Tushar Agarwal, Karlo Basioli, Theotime Combes, Emilio Cota, Minos Papadopoulos, Tudor Pipernea, Yarong Mu, Qi Zhou, Yaoshiang Ho, Aleksey Vlasenko, Alexander Brauckmann, Ziyin Huang, Jack Oh, Jerry Hsu, Tanvi Bajpai, Jialei Chen, Wenjun Gu, Chintan Shah, Chris Chan, Sayce Falk, Chris Jones, Rob Torres, Kat Ko, Lee Howes, Freya Azad, Andi Gavrilescu, Joe Pamer, Bill JiaAbstract
TorchTPU is an open-source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the-box model execution and TPU migration while delivering industry-competitive inference and training performance through eager execution optimizations, this http URL integration, and native hardware acceleration.
Abstract (translated)
URL
https://arxiv.org/abs/2610.24161
PDF
-
How Data Scales in Agentic Reinforcement Learning: Laws and Synthesis Strategies
2026-10-20 01:59:17Bowei He, Yankai Chen, Xiaokun Zhang, Changjiang Han, Ye Yuan, Chong Li, Xue LiuAbstract
Pretraining scaling laws treat training data as a one-dimensional quantity: a token count. Agentic reinforcement learning (RL) inherits this scalar abstraction, but modern synthesis pipelines now generate \textit{tasks}, \textit{environments}, and \textit{trajectories} independently and at very different costs, turning data into a structured design space. This raises a question pretraining never had to answer: given a synthesis budget, \textit{which axis} should be scaled? We address this question end-to-end with three tightly coupled contributions. First, we decompose agentic data into three cardinality axes (task, environment, trajectory), plus a reward-density control variable, and fit per-axis scaling relations on different large language models (LLMs) across math+code, SWE, web, and UI/tool application domains. Across all model sizes, model families, domains, and RL algorithms (PPO and GRPO) we measured, the scaling exponents order as environment $>$ task $>$ trajectory. When environment diversity is decomposed into world and environment-structure diversity, its advantage is mainly driven by the former, while the latter is just comparable to the task diversity. Second, we use the fitted laws as a common ruler to price synthesis methods, capturing each method's efficiency as a per-axis discount factor and identifying critical synthetic-to-real ratios at which collapse begins; these ratios differ by an order of magnitude across axes. Third, we cast budget allocation as a constrained optimization under our laws and discounts, deriving a Chinchilla-style compute-optimal recipe and validating it at held-out scale. The empirical results demonstrate that following the recipe yields up to $1.7\times$ compute efficiency over the dominant ``scale-the-trajectories'' practice.
Abstract (translated)
URL
https://arxiv.org/abs/2610.23946
PDF
