arXiv subjects

Berkcan Kapusuzoglu

Publications and source records attributed to Berkcan Kapusuzoglu.

13 recordsLinked to original sources

arXiv2026

When Load-Balancing Goes Too Far: Expert Pruning in Over-Dispersed Mixture-of-Experts Models

Expert pruning reduces the memory and serving cost of Mixture-of-Experts (MoE) models by removing low-importance experts identified by the router, assuming router probabilities provide a reliable importance signal. We observe that this assumption breaks down under over-dispersed routing, a regime associated with aggressive load-balancing during training, in which tokens are distributed nearly uniformly across experts and importance signals collapse. In this regime, perplexity does not predict downstream task accuracy: on gpt-oss-20B, the lowest-perplexity pruning configuration yields the worst mathematical reasoning, while the highest-perplexity configuration preserves it. This does not occur under standard routing (e.g., Mixtral-8x7B-Instruct), where perplexity and accuracy degrade together. Pruning under over-dispersed routing also exposes a capability trade-off in which no single scoring metric dominates: activation-aware scoring preserves mathematical reasoning but severely degrades knowledge-intensive science (an 18-point gap on GPQA), whereas frequency-based scoring exhibits the reverse. We propose Minimax Expert Score Allocation (MESA), a domain-aware method that iteratively boosts importance scores for experts serving whichever domain is currently worst-affected, minimizing worst-case domain degradation rather than average accuracy. At 25% expert pruning MESA achieves the smallest worst-case degradation across domains, outperforming activation-aware baselines on 7 of 11 benchmarks at a correspondingly reduced memory footprint, and it generalizes to gpt-oss-120B, Gemma-4-26B-A4B, and OLMoE-1B-7B. Our results indicate that over-dispersed routing is a qualitatively distinct pruning regime in which standard assumptions fail, and that recognizing it is a prerequisite for principled expert pruning of load-balanced MoE models.

cs.LG↗

arXiv2026

Process Optimization Under Uncertainty for Improving the Bond Quality of Polymer Filaments in Fused Filament Fabrication

This paper develops a computational framework to optimize the process parameters such that the bond quality between extruded polymer filaments is maximized in fused filament fabrication (FFF). A transient heat transfer analysis providing an estimate of the temperature profile of the filaments is coupled with a sintering neck growth model to assess the bond quality that occurs at the interfaces between adjacent filaments. Predicting the variability in the FFF process is essential for achieving proactive quality control of the manufactured part; however, the models used to predict the variability are affected by assumptions and approximations. This paper systematically quantifies the uncertainty in the bond quality model prediction due to various sources of uncertainty, both aleatory and epistemic, and includes the uncertainty and the model discrepancy in the process parameter optimization. Variance-based sensitivity analysis based on Sobol indices is used to quantify the relative contributions of the different uncertainty sources to the uncertainty in the bond quality. A Gaussian process (GP) surrogate model is constructed to compute and include the model discrepancy within the optimization. Physical experiments are conducted for calibration and validation of the physics model and also for validation of the optimum solution. The results show that the proposed formulation for process parameter optimization under uncertainty results in high bond quality between adjoining filaments of the FFF product.

cs.CE↗

arXiv2026

Multi-Objective Optimization Under Uncertainty of Part Quality in Fused Filament Fabrication

This work presents a data-driven methodology for multi-objective optimization under uncertainty of process parameters in the fused filament fabrication (FFF) process. The proposed approach optimizes the process parameters with the objectives of minimizing the geometric inaccuracy and maximizing the filament bond quality of the manufactured part. First, experiments are conducted to collect data pertaining to the part quality. Then, Bayesian neural network (BNN) models are constructed to predict the geometric inaccuracy and bond quality as functions of the process parameters. The BNN model captures the model uncertainty caused by the lack of knowledge about model parameters (neuron weights) and the input variability due to the intrinsic randomness in the input parameters. Using the stochastic predictions from these models, different robustness-based design optimization formulations are investigated, wherein process parameters such as nozzle temperature, nozzle speed, and layer thickness are optimized under uncertainty for different multi-objective scenarios. Epistemic uncertainty in the prediction model and the aleatory uncertainty in the input is considered in the optimization. Finally, Pareto surfaces are constructed to estimate the tradeoffs between the objectives. Both the BNN models and the effectiveness of the proposed optimization methodology are validated using the actual manufacturing of the parts.

cs.CE↗

arXiv2026

Multi-Level Bayesian Calibration of a Multi-Component Dynamic System Model

This paper proposes a multi-level Bayesian calibration approach that fuses information from heterogeneous sources and accounts for uncertainties in modeling and measurements for time-dependent multi-component systems. The developed methodology has two elements: quantifying the uncertainty at component and system levels, by fusing all available information, and corrected model prediction. A multi-level Bayesian calibration approach is developed to estimate component-level and system-level parameters using measurement data that are obtained at different time instances for different system components. Such heterogeneous data are consumed in a sequential manner, and an iterative strategy is developed to calibrate the parameters at the two levels. This calibration strategy is implemented for two scenarios: offline and online. The offline calibration uses data that is collected over all the time-steps, whereas online calibration is performed in real-time as new measurements are obtained at each time-step. Analysis models and observation data for the thermo-mechanical behavior of gas turbine engine rotor blades are used to analyze the effectiveness of the proposed approach.

cs.CE↗

arXiv2026

Adaptive surrogate modeling for high-dimensional spatio-temporal output

This paper develops an adaptive surrogate modeling method for problems with very high-dimensional spatio-temporal outputs. The analysis of spatio-temporal multi-physics systems is computationally expensive and consists of a large number of inputs and outputs. Surrogate models are often constructed to replace the physics-based model to achieve computational efficiency in analyses such as uncertainty quantification and optimization that require many function calls. In order to address the challenge introduced by the high dimensionality of spatio-temporal output, a dimension reduction method is first employed to map the high-dimensional output to a low-dimensional latent space. This is followed by the construction of the surrogate model in the low-dimensional space. The prediction error in the original space, which includes both the reconstruction error and surrogate model error, is evaluated using different error metrics. Based on the prediction accuracy of the surrogate model, new training points are identified for adaptive improvement of the surrogate model. We present a novel adaptive sampling technique that combines exploration and exploitation to improve the surrogate model accuracy with the fewest possible runs of the expensive physics-based model. Thermo-mechanical analysis of a gas turbine engine blade is used to analyze the effectiveness of the proposed method.

cs.CE↗

arXiv2026

Physics-Informed and Hybrid Machine Learning in Additive Manufacturing: Application to Fused Filament Fabrication

This article investigates several physics-informed and hybrid machine learning strategies that incorporate physics knowledge in experimental data-driven deep-learning models for predicting the bond quality and porosity of fused filament fabrication (FFF) parts. Three types of strategies are explored to incorporate physics constraints and multi-physics FFF simulation results into a deep neural network (DNN), thus ensuring consistency with physical laws: (1) incorporate physics constraints within the loss function of the DNN, (2) use physics model outputs as additional inputs to the DNN model, and (3) pre-train a DNN model with physics model input-output and then update it with experimental data. These strategies help to enforce a physically consistent relationship between bond quality and tensile strength, thus making porosity predictions physically meaningful. Eight different combinations of the above strategies are investigated. The results show how the combination of multiple strategies produces accurate machine learning models even with limited experimental data.

cs.LG↗

arXiv2026

Information fusion and machine learning for sensitivity analysis using physics knowledge and experimental data

When computational models (either physics-based or data-driven) are used for the sensitivity analysis of engineering systems, the sensitivity estimate is affected by the accuracy and uncertainty of the model. This paper considers global sensitivity analysis (GSA) for situations where both a physics-based model and experimental observations are available, and investigates physics-informed machine learning strategies to effectively combine the two sources of information in order to maximize the accuracy of the sensitivity estimate. Two representative machine learning (ML) techniques are considered, namely, deep neural networks (DNN) and Gaussian process (GP) modeling, and two strategies for incorporating physics knowledge within these techniques are investigated, namely: (i) incorporating loss functions in the ML models to enforce physics constraints, and (ii) pre-training and updating the ML model using simulation and experimental data respectively. Four different models are built for each type (DNN and GP), and the uncertainties in these models are included in the Sobol indices computation. The DNN-based models, with many degrees of freedom in terms of model parameters and training options, are found to result in smaller bounds on the sensitivity estimates when compared to the GP-based models. The proposed methods are illustrated for additive manufacturing and lake temperature modeling examples.

cs.CE↗

arXiv2026

Structured Thoughts For Improved Reasoning And Context Pruning

Large language models (LLMs) excel at generating long chains of thought, but long reasoning traces are often verbose and memory-inefficient. In this work, we introduce Structured Thoughts, a framework that organizes reasoning into alternating and blocks: captures exploratory scratch work, while contains the distilled conclusion of that step. We construct a dataset of structured thoughts by segmenting reasoning traces into blocks and prompting an LLM to summarize each step into its corresponding . Fine-tuning pretrained foundation models on this reformatted data produces models that adopt the structured reasoning style, leading to performance gains of up to 8.08\% on reasoning benchmarks compared to standard SFT. The explicit structure also enables context pruning: after each / pair, the can be pruned, allowing the model to retain conclusions without keeping the full scratch work in the context. A proof-of-concept pruning implementation achieves an average of 85\% memory / context savings with an 8.67\% performance drop across mathematical tasks.

cs.CL↗

arXiv2026

On the Policy Gradient Foundations of Group Relative Policy Optimization: Credit Assignment, Gradient Sparsity, and Rank Collapse

Group Relative Policy Optimization (GRPO) eliminates the learned critic in PPO by using the mean reward of grouped rollouts as a baseline. We provide a rigorous derivation of GRPO from first principles of the policy gradient theorem, revealing a fundamental credit assignment failure: under output-only reward, every token in a rollout receives identical advantage, collapsing token-level credit to a single scalar. We prove this induces gradient sparsity that intensifies over training, and demonstrate empirically via SVD analysis of GRPO gradients on Nemotron-4B/GSM8K that the gradient matrix has effective rank $\approx$ 2 regardless of group size $R \in \{2, 4, 8\}$. We formalize this as an intrinsic rank-2 structure arising from the zero-sum constraint on advantages and derive conditions under which GRPO's baseline is optimal. Our results characterize when GRPO's simplicity is theoretically justified and identify the credit assignment bottleneck as the key limitation for multi-step reasoning.

cs.LG↗

arXiv2026

CoT-Guard: Small Models for Strong Monitoring

Monitoring the chain-of-thought (CoT) of reasoning models is a promising approach for detecting covert misbehavior (i.e., hidden objectives) in code generation tasks. While large models (GPT-5, Gemini-3-Flash) can serve as effective CoT monitors, they are expensive to deploy due to the lengthy reasoning traces and high API cost, emphasizing the need for smaller, cheaper alternatives. Nevertheless, we find that current small models (4B--8B) struggle to detect hidden objectives despite access to the CoT, frequently misattributing them as part of the user query. To address this, we propose a post-training pipeline combining supervised fine-tuning (SFT) and reinforcement learning (RL), where SFT narrows the gap for in-domain tasks by distilling detection behavior from stronger monitors, and RL on hard and subtly crafted hidden objectives helps the model generalize to out-of-domain monitoring tasks. To validate this generalization, we evaluate under a realistic threat model motivated by practical supply-chain attacks, where the adversary is a third-party LLM router injecting hidden objectives into code-generation requests through either prompt manipulation or code manipulation attacks. To push beyond objectives that large monitors already saturate, we also introduce four new challenging tasks even for strong monitors. Finally, we introduce CoT-Guard, a 4B-parameter monitor that demonstrates superior generalization performance under both prompt and code manipulation attacks, achieving a G-mean^2 (i.e., TNR x TPR) of 75% and outperforming GPT-5.4 (56%), GPT-5-mini (41%), and Qwen3-32B (54%), while closing the gap to Gemini-3-Flash (83%). These results demonstrate that CoT-Guard provides a practical and cost-effective user-side defense, substantially improving hidden-objective detection while avoiding the deployment cost of large monitors.

cs.CR↗

arXiv2025

SPEAR-MM: Selective Parameter Evaluation and Restoration via Model Merging for Efficient Financial LLM Adaptation

Large language models (LLMs) adapted to financial domains often suffer from catastrophic forgetting of general reasoning capabilities essential for customer interactions and complex financial analysis. We introduce Selective Parameter Evaluation and Restoration via Model Merging (SPEAR-MM), a practical framework that preserves critical capabilities while enabling domain adaptation. Our method approximates layer-wise impact on external benchmarks through post-hoc analysis, then selectively freezes or restores transformer layers via spherical interpolation merging. Applied to LLaMA-3.1-8B for financial tasks, SPEAR-MM achieves 91.2% retention of general capabilities versus 69.7% for standard continual pretraining, while maintaining 94% of domain adaptation gains. The approach provides interpretable trade-off control and reduces computational costs by 90% crucial for resource-constrained financial institutions.

cs.CL↗

arXiv2025

Optimizing Reasoning Efficiency through Prompt Difficulty Prediction

Reasoning language models perform well on complex tasks but are costly to deploy due to their size and long reasoning traces. We propose a routing approach that assigns each problem to the smallest model likely to solve it, reducing compute without sacrificing accuracy. Using intermediate representations from s1.1-32B, we train lightweight predictors of problem difficulty or model correctness to guide routing across a pool of reasoning models. On diverse math benchmarks, routing improves efficiency over random assignment and matches s1.1-32B's performance while using significantly less compute. Our results demonstrate that difficulty-aware routing is effective for cost-efficient deployment of reasoning models.

cs.LG↗

arXiv2025

Critique-Guided Distillation for Robust Reasoning via Refinement

Supervised fine-tuning with expert demonstrations often produces models that imitate outputs without internalizing the reasoning processes needed for robust generalization. While critique-based approaches show promise, training models to generate critiques directly, such as Critique Fine-Tuning (CFT), can lead to output-format drift and degradation of general capabilities. We propose Critique-Guided Distillation (CGD), a training framework that decouples critique consumption from critique generation. During fine-tuning, the student is trained to refine flawed responses conditioned on teacher critiques. CGD treats critiques as a \textit{training-time-only} supervision signal, encouraging internalization of error-aware reasoning: critiques guide learning but are absent at inference. Controlled ablations confirm that these reasoning gains are directly driven by the specificity and relevance of the teacher's feedback. Across five model families, CGD consistently outperforms CFT and standard distillation on mathematical reasoning benchmarks, yielding 7\% average improvements and gains of up to +15.0\% on AMC23 and +12.2\% on MATH-500. On challenging competition problems such as AIME24 and AIME25, CGD achieves substantially higher Pass@1 and stronger performance at low Pass@k, indicating improved reasoning quality per sample. Importantly, CGD preserves general instruction-following capabilities where CFT degrades significantly ($-$21.3\% on IFEval). These results position CGD as a practical and compute-efficient intermediate training paradigm for reasoning-centric tasks without introducing architectural inference-time overhead.

cs.CL↗