The paper proposes a structural method using decision tree rulesets and multiple complementary metrics to detect concept drift in evolving malware families, finding that fixed-interval windowing with…

A hybrid active-online learning framework is proposed for label-efficient concept drift adaptation in optical network failure detection, achieving high accuracy and AUC scores with minimal samples and…

The paper proposes a cost-aware, adaptive maintenance framework using Reinforcement Learning (RL) and self-supervised learning to mitigate performance degradation (concept drift) in Android malware de…

This study longitudinally evaluates the adversarial robustness of Android malware detection systems over a decade, finding that temporal separation significantly degrades robustness due to concept dri…

EvoGens is an evolution-inspired framework that treats scientific idea generation as an evolutionary search, significantly boosting the novelty and diversity of generated research ideas compared to ex…

The paper introduces IdeaGene-Bench, a benchmark for scientific lineage reasoning and idea generation, which includes 1,961 golden lineage traces, 1,085 curated Idea Genome objects, and 920 pairwise G…

This paper unifies the fragmented field of Tree-of-Thoughts (ToT) reasoning by mapping LLM-based search processes onto a formal taxonomy derived from classical heuristic search theory.

The paper introduces Drifting Preference Optimization (DrPO), an efficient online method for preference finetuning one-step text-to-image generators that avoids complex gradient calculations and model…

DRIFT proposes a novel framework that efficiently optimizes LLMs for multi-turn interactions by decoupling rollout from optimization, allowing the use of weighted supervised fine-tuning to match the p…

The paper proposes Neighbor-Aware Localized Concept Erasure (NLCE), a training-free framework that effectively removes specific concepts from text-to-image models while minimizing the unintended degra…

This paper characterizes the convergence pattern of chain-of-thought reasoning models and finds that convergence fate is partially encoded in intermediate representations.

The paper proposes Orthogonal Concept Erasure (OCE), a novel multiplicative parameter update method that achieves precise concept erasure in diffusion models while independently preserving overall gen…

DriftQL introduces a novel, efficient offline RL method that combines a drift-based behavioral regularizer with critic-driven policy improvement, achieving state-of-the-art performance while maintaini…

The paper introduces SafetyDrift, a predictive model that forecasts when AI agents will violate safety protocols by analyzing the cumulative risk across sequences of individually safe actions.

The paper identifies 'memory-induced tool-drift,' a systematic vulnerability where personality biases stored in an LLM agent's memory silently corrupt tool-calling decisions, even when those biases ar…

CB-SLICE is a novel concept-based method for discovering model error slices that leverages Concept Bottleneck Models (CBMs) to provide fine-grained, faithful explanations directly linked to the root c…

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…

The paper evaluates the gap between LLM-generated research ideas and human research ideas by building an evaluation framework and observing a distributional gap.

The paper compares anchorless methods for diversifying LLM-generated idea pools against traditional anchor-dependent methods, finding that semantic direction stratification offers the best balance of…