This paper introduces Stemma, a black-box LLM fingerprinting method that abstracts away surface-form variation and measures the inheritance of decision regions for reliable provenance testing.

This paper proposes a new training objective for sample-based generative models that considers decision maker's cost structure.

The paper introduces PRAXIS, a novel algorithm that efficiently approximates the computation of 'Rashomon sets' for decision trees, significantly reducing memory and runtime complexity.

The paper analyzes the entropy dynamics of Chain-of-Thought (CoT) reasoning, identifying a transition from an exploratory Uncertainty Region to a stable Confidence Region, which enables superior early…

This paper extends gradient boosting to functions of vector inputs using a simple algorithm with histogram-based decision trees.

The paper introduces Trust Region On-Policy Distillation (TrOPD), a robust method that stabilizes the on-policy distillation of large language models by restricting training to regions where teacher s…

This paper proposes a Security Decision Support System that recommends security control sub-families using a curated dataset and multi-agent model.

The paper introduces Entropy-Cut Metropolis-Hastings, an efficient sampling method that uses next-token entropy to identify and resample from critical decision points in a reasoning trace, significant…

The paper proposes GRiD, a novel framework that uses a two-phase training strategy (supervised pre-training and RL fine-tuning) to discover complex, graph-like rules for knowledge graph reasoning, ove…

This paper introduces a method to automatically determine the optimal learning period ($ au$) for the Random Gradient hyper-heuristic, enabling it to optimally solve Pseudo-Boolean Problems without ma…

This paper synthesizes over 150 scattered studies and reports to provide the first comprehensive primer on post-training reasoning data, organizing the field around data objects, utility, construction…

The paper introduces ForeSci, a novel benchmark that evaluates LLM agents' ability to make forward-looking research judgments using only historical evidence, finding that explicit evidence organizatio…

This paper demonstrates that a complex deep reinforcement learning policy for power grid control can be successfully distilled into a lightweight, auditable decision tree and random forest surrogate t…

The paper proves XNLP-hardness of Directed Edge Geography and Undirected Edge Geography when parameterized by pathwidth, and shows their fixed-parameter tractability when parameterized by treewidth an…

The paper introduces AdaPrefix-GRPO, a method that adjusts the amount of reference solution assistance during training to improve the success rate and accuracy of Group Relative Policy Optimization (G…

This paper introduces MCTS-Guided Group Relative Policy Optimization (M-GRPO) to enhance LLM spatial reasoning by improving the decomposition of complex tasks into optimal sub-tasks.

This paper proposes a hierarchical analytical framework to characterize region-level latency differences in Low-Earth orbit satellite Internet using Starlink RTT measurements.

The paper introduces ClinPivot, a benchmark that tests whether clinical models can correctly adjust treatment decisions when new patient context constraints are introduced, finding that strong medical…

The paper proposes a system that operationalizes cybersecurity governance frameworks by integrating them with attack-path modeling and Deep Reinforcement Learning to generate practical, resource-const…